Final Design Report
Second Wind
Final Design
Dec. 8, 2009
Josh Dowler
Caleb Meeks
John Snyder
1
Table of Contents
Requirements Specification...………………………………………….…………………………….………………….…..…….……….….3
Final Design…………………………………..……………………………………………………………………………………………….…..…….5
Block Diagram……………………………………………………………..…………………………………………………..….….....6
Organization and Management.………………………………………………..………………………….…..……..………..7
Generator Motor Selection.…….………………………………………………………..…………………...………...……..…8
Sprockets and Chains……………..…………………………………………………………..…………………………..………....9
Motor and Gearing Calculations……...………………………………………………..…………….…….………..………..11
Freewheel and Return Mechanisms..………………………………………………..…………….……….………...…….12
Generation Stress Concentration Points……………………………………………………..….……….………...……..14
Generation System Diagrams….…………………………………………………..………………………………….…..……15
Charge Controller Design ………………………….…………..……………….……………………….….………….………..20
Microprocessor Design………………………….…………..……………….…………………………….….................…..25
Software Design…………………………..………….………..……………….……………………………………………….…...27
Motor Controller Design………………………….………..……………….……………………………………………….…...30
Electrical Components………………………….…………..……………….…………………………………………..…...…..31
Electronically Assisted Design…………………………………………………………..……………………………………….33
Kite Retraction System Design………………………..………………………………..…………………………..………….43
Budget Analysis.………………..…….…………………….…………………………………..…………………………...….……45
Updated Budget………………..………………………………………………………………..…………………..………….……46
Schedule Analysis.……………..………………………………..…………………………………..……………..………….……47
Gantt Chart – Fall 2009…………………………………………………………………………..………………..…….…………48
Gantt Chart – Spring 2010…………………………………………………………………..…………………………………….49
Appendices…………………………………………………………………………………………………..…………………………….…......….50
MAX15046 DC-DC Converter………………………………………………………………………………………..…….………A
PIC18F4550 Microprocessor……………………………………………………………………………………………………….B
TLE5205-2 Motor Controller……………………………………………………………………………………………………….C
2
Kite Wind Generator
Requirements Specification
Overview:
Our team will design and prototype a kite wind generator. The generator will produce
electrical power from the drag force applied to the kite by wind. The kite will be autonomously
guided by a microprocessor to perform the gliding maneuvers necessary to produce power. A kite
wind generator would be useful for generating power on large scale agricultural farms, in remote
locations for disaster relief or military, or as a part of a larger wind farm.
Problem Statement:
Due to pollution and depletion of traditional energy sources there is a need to generate
power from renewable energy sources. Wind is the second most abundant energy resource, next
to solar energy, that can be harnessed to generate power. Kite wind generation is more effective
than conventional turbines in gathering the energy from the wind for two reasons. First, the kite
can reach much higher altitudes than turbines, where the wind is more reliable and strong.
Second, kites can cover more area in the sky and therefore use more of the energy than a
stationary turbine can. This technology could allow individuals to become energy self-sufficient and
it could also be used in large scale projects as wind farms that produce high power.
Operational Description:
The kite wind generation unit will produce power based on the drag force produced by the
kite in flight and the amount of line pulled, which will be connected to a generator, over time.
When the kite has reached its maximum height the kite orientation will be changed to reduce its
drag coefficient, and the kite will be retracted using much less power than is generated from the
pull up. The kite will run autonomously in winds of 10 to 45 kilometers per hour. When the wind
speeds are too high the kite will be retracted to prevent damage to the system. If the wind speeds
are too low the kite will be retracted. The system will also have a user interface that displays the
length of line released, and power generation. The user will also have options for three different
modes of operation for the kite; deploy, sustain, and retract.










Technical Requirements:
System will initially supply its own power to initiate energy generation and then store excess generated
power
If power generation is not sufficient to generate excess power the kite will be retracted and the user
interface will run off of stored power
System will generate at least 500 watt hours DC within 10 hours
Kite system will be able to generate power in winds from 10 - 45 kilometers per hour
Setup, including kite deployment, should take no more than 30 minutes
Power generation should occur within five minutes of kite deployment
System must have deploy, sustain, and retract modes of operation
Autonomous control of each mode (deploy, sustain, retract)
User interface to enable user to specify modes of operation (deploy, sustain, retract) and show user
length of line released within one meter and power generated within 20 watts
Must be able to sense length of line released within one meter and power generation within 20 watts
3








1.
2.
3.
4.
5.
6.
7.
System will be able to fit through a standard door frame, with width of one meter and height of two
meters
Design Deliverables:
User manual
Drawings and schematics with analyses
Kite generator unit
User interface
Parts list with associated costs
Test report
Final technical report
System Test Plan
Kite stays aloft in winds of 10 - 45 kilometers per hour
10 minutes of autonomous flight and power generation in winds of 10 - 45 kilometers per hour
Generation of 500 watt hours DC within 10 hours
The electrical system will have a fail safe mechanism that will enable in case of a power surge
Kite retraction of less than 10 minutes in winds of 10 - 45 kilometers per hour
Shows accurate value for length of line released by comparing it with a tape measure within one meter
Shows accurate value for power generation within 20 watts by using current and voltage measurements
using a multimeter
Implementation Consideration:
Follow FAA regulations part 101, subparts A and B: no flight between sunset and sunrise, a
letter of intent to fly the kite above 150 feet sent to the nearest FAA ATC facility, a 100m radius of
land without obstruction around base, set in an area five miles away from an airport, land must
have ground visibility greater than 3 miles, and the kite line must have streamers at 50 foot
intervals above 150 feet that are visible for one mile. The leads for the generator and battery will
be covered to prevent shock. Sprockets and chains are part of the design and could propose some
safety issues.
4
Final Design
5
Block Diagram
6
Organization and Management
John Snyder – John is a senior computer engineering student, with a 50/50 electrical and engineering
and computer science split. He will be working with programming the microprocessor to get it
to work with the motor controller, kite controls system, and the user interface. He will also be
working on the charge controller to prevent it from overcharging and surge protection for the
power supply. He will also be working with different sensors to provide information for the
system.
Josh Dowler – Josh is a senior mechanical engineering student, and is the project leader. He will be in
charge of converting the tension provided by the kite behavior and turning it into electric
power. He will be working with the generator motor and a freewheel mechanism to allow the
kite to retract without affecting the generator and selecting gear ratios as necessary. As project
leader, he will be in charge of managing the budget, overseeing all project happenings, and
reviewing documentation.
Caleb Meeks – Caleb is a senior mechanical engineering student. He will be in charge of working with
the controls system and kite behavior. He will construct and work closely with John on the
electrical and mechanical aspects of the controls system. The controls system will also link with
the power generation processes, and therefore Caleb and Josh will be working to integrate their
systems.
All team members will contribute equally to any documentation that will be presented,
including reports and oral presentations. Each team member will be in charge of maintaining
their notebooks and doing research on their respective parts outside of group meeting times.
Team members are required to attend team meetings unless they notify the other team
members about their absence.
7
Generator Motor Selection
Ideal motor criteria:
1) Reversible - motor can act as a generator
2) DC - no AC/DC inverter needed
3) Permanent Magnet - powerful
4) Brushless - less friction loss
5) Low rated RPM - less gearing required
6) Continuous Duty - made to run continuously
Based on some basic lift and drag force
equations (
,
) tension force
from the kite onto the system was calculated. In these
equations; p is air density, ν is the wind speed, A is the
characteristic area of the kite, and Cl and Cd are the lift and
drag coefficients, respectively. The tension force was then
Motor Torque Gear Kite Tension Motor Power
Ratio
in*lbf (N*m)
lbf (N)
(watts)
1.0 (0.113)
7.4
3.5 (15.6)
30.76
2.0 (0.246)
7.0
7.0 (31.1)
61.53
3.0 (0.369)
7.0
10.5 (46.7)
92.29
4.0 (0.492)
7.0
14.0 (62.3)
123.05
5.0 (0.615)
7.0
17.5 (77.8)
153.82
6.0 (0.738)
7.0
21.0 (93.4)
184.58
7.0 (0.861)
7.0
24.5 (109.0)
215.34
8.0 (0.984)
7.0
28.0 (124.6)
246.11
9.0 (1.107)
7.0
31.5 (140.1)
276.87
10.0 (1.230)
7.0
35.0 (155.7)
307.63
11.0 (1.353)
7.0
38.5 (171.3)
338.40
12.0 (1.476)
7.0
42.0 (186.8)
369.16
13.0 (1.599)
7.0
45.5 (202.4)
399.92
14.0 (1.722)
7.0
49.0 (218.0)
430.68
15.0 (1.845)
7.0
52.5 (233.5)
461.45
Assuming a 4" diameter spring return mechanism
Table 1
converted into a torque value
based on
assumed values for sprockets used to gear the system.
The torque transferred to the motor varied based upon
the assumed values for the sprockets, so an Excel spreadsheet was used to calculate various
possible torque values that could be transferred to the motor based upon the gearing ratio and
the speed of the kite pulling out (Figure 2).
Based upon the calculated torque that could be supplied to the system our system will
be able to overcome the torque for a 350 watt motor. Many inexpensive brushed motors are
available with power outputs around 350 watts. One brushless motor was found that had a
power output in the range of 350 watts. However, this motor is not reversible and thus cannot
be used as a generator
The chosen motor is priced at $47.91 from Monster Scooter Parts shown in table 1. The motor
has an 11 tooth sprocket for #25 roller chain. The system will need a form of gearing to reach
the rated motor speed.
Table 2
8
Sprockets and Chains
The design of the gearing system began during the generator motor selection phase due
to the dependency between the two systems. The force coming in from the lines needs to be
geared up to supply the rated RPMs of the motor.
The use of belts and pulleys were
considered along with chains and sprockets.
However, it soon became apparent that
belts and pulley are much more expensive
and difficult to find in varying sizes. The
efficiencies of chains and belts are both
above 95%, so this was not a factor. A
decision matrix was used to analytically
make this decision. Chains are obviously
the better choice for this system.
Table 3
The initial design used only
one shaft on which a large gear
would sit with the return mechanism
and directly connect to the motor
(Figure 1). However, it was soon
determined that the addition of an
intermediate shaft was necessary to
allow for proper use of the freewheel
mechanism when connected to the
large sprocket (Figure 2). The
intermediate shaft, which contains
two sprockets and a flywheel,
connects the drive shaft and the
motor sprocket. This means that the
first shaft contains the spring return
mechanism and the large sprocket
connected to the freewheel system.
This configuration, in conjunction
with the flywheel, allows the
intermediate shaft, and therefore the
motor, to continue spinning between
the oscillatory pull of the kite on the
system.
Figure 1
Figure 2
9
The generator motor uses #25 roller chain, therefore the sprockets must also use #25
roller chain. The tension of the kite lines at maximum wind speed is 50.97 lbs (226.75 N), as
calculated from the lift and drag force equations stated earlier. The working load of #25 roller
chain is 140 lbs (622.75 N) which gives a factor of safety of 2.75 for the chains. The designed
system gearing ratio will be 7.4:1, which will give a balance between the kite speed required
and the torque the kite tension force will generate. The Excel data and graph is shown on the
following page in Figure 3 and Table 4.
A fully assembled rear wheel assembly comes with the large sprocket combined with the
freewheel, and axle size needed as well as a wheel and tire that is not needed. Two sprockets
(one 14 tooth and one 16 tooth), a steel bar of ½ inch diameter, and two bearings for the
intermediate shaft will be purchased.
The flywheel does not have a specific design as of yet. From research it has been found
that the heavier the flywheel the smoother the operation of the motor, as long as the driving
force is able to overcome the inertia of the flywheel. The design portion on flywheels in the
machine design text book is not clear in the actual process for designing the flywheels. There
are also no online resources that use an engineering approach to designing flywheels.
I will continue to do more research on the design of flywheels.
10
Motor and Gearing Calculations
The variables in
Table 4 are defined in
Figure 1 on page 12.
R2 is varied to find the
radius of the large
sprocket needed to
attain a balance
between kite speed
(Ks) and force of the
kite (Fk).
A gear ratio
that balances kite
speed and force of the
kite can be determined
using the graph of
Figure 3. This gives a
gear ratio R2:R1 of
approximately 7.4:1.
Thus the kite will need
to move around 6.5
feet per oscillation of
the system based on
Table 4.
Table 4
Force of the Kite vs. Kite Speed
16
Kite Speed (ft/s)
14
12
Desired Gear Ratio ≈ 7.4:1
10
8
6
4
2
0
10
20
30
40
Force of the Kite Fk (lbs)
Figure 3
11
50
60
Freewheel and Return Mechanism
The freewheel mechanism is a vital part of the
generation system design. Figure 4 shows how a basic
freewheel mechanism works. A freewheel allows free rotation
in one direction, but when spun in the opposite direction the
system engages and allows energy transfer. This means that
the kite string, when pulled out, will engage the freewheel and
transfer energy into the shafts and motor. However, when the
kite is in the return stage of its oscillatory motion, due to the
spring return mechanism, the kite string and the motor can
move independently.
This system will use a freewheel that is already
Figure 4
attached to a sprocket and axle on a rear wheel assembly of
an electric scooter. There are no force or life specifications on the freewheel that we will be
using. However, it is assumed that this system will undergo far less stress compared to a high
powered electric scooter that the part is made for.
The return mechanism is the system that will be used to retract the kite during each
oscillation of the kite through its figure eight pattern. The return mechanism consists of a spring
that will store some of the energy of the kite into potential spring energy when the kite is on its
outward pull. The potential spring energy will then be released when the kite is on the outside
of its figure eight flight pattern, which will be discussed later, and the kite will be retracted to
its original position.
Kite string
String Connecting
Spring and Variable
Diameter Spools
The spring return mechanism is
more difficult to design due to the need
for a constant kite pull out length at
varying wind speeds. R3 (Figure 2)
needs to remain constant at 2 inches
due to the gearing ratio already being
selected. The values for the force of
the kite are already known as shown in
Table 5. The cheapest and easiest way
to allow the kite to have a constant pull
out length at varying wind speeds is to
vary the spring constant ‘k’ of the
spring. Under the budget of this
Shaft
Attached Spring
Variable Diameter
Spools for Spring
Return Mechanism
Figure 5
12
project the best way to do that is to have the
operator manually adjust the spring for the
current relative wind speed. Setting the
torque of the kite on the shaft equal to the
torque a spring on the shaft allowed iteration
to find a ‘k’ and spring extension ‘∆X’ that
could be manufactured and purchased.
The springs found for minimum and
average winds happened to be able to use the
same radius ‘R0’ that is defined in Figure 5 as
the variable diameter spools for spring return
mechanism. The spring found for the
maximum wind speed required a smaller R0 to
allow for the necessary six feet as seen on the
motor and gearing calculations page. The use
of a linear spring was determined by the wide
variety of linear springs over torsion springs on
the market. Furthermore, linear springs
generally have a smaller ‘k’ value which will
allow the kite to meet its displacement
requirements during oscillation.
The springs being used by this system are
from W.B. Jones Spring Company Inc. W.B. Jones
offers quick and easy purchase of a number of
stock linear springs. Table 6 shows the
specifications for the three springs chosen for
minimum, average, and maximum wind speed
values.
Table 5
Table 6
13
Generation System Finite Element Analysis
Finite Element Analysis was completed on the shaft that would have the most likelihood to
fail. This is due to the large load of the flywheel on the shaft as well as the two intermediate
sprockets. There is a force of 50 lbs (N) in the center of the shaft, where there would actually be a
weight of 25 lbs (N). There are also two torsion forces where the intermediate sprockets will go
with a value of 22 lb*in (), where there would actually be only as much as 11 lb*in of torque.
Figures 6, 7, and 8 below represent the deflection, stress, and factor of safety of the beam with
exaggerated deflections. The deflections are well below one thousandth of an inch. This proves that
the shaft will easily support the forces acting on it.
Figure 6
Figure 7
Intermediate Sprocket Locations
Flywheel Location
Figure 8
14
Power Generation System Diagram
15
Power Generation System Diagram (Cont.)
16
Power Generation System Diagram (Cont.)
17
Power Generation System Diagram (Cont.)
18
Power Generation System Diagram (Cont.)
19
Charge Controller Design Development of a proper charge controller for our application required the consideration of many different factors. This begins with the motor that is generating our power. We have a 24V, 350W DC motor as our generator. This means the charge controller needs to be able to handle high voltage (up to 24V) and high current inputs (up to 14A). It then needs to regulate those high inputs into a manageable power source to charge a battery. The battery chosen is a 12V 26Ah lead acid battery. This battery will allow a capacity of 312 watt hours. The entire system needs to prove it can generate 500 watt hours within a ten hour period. Since storing all of this energy to the battery is not possible, the current and voltage generated will be monitored over that time to compute the average energy production. Since constant current and voltage generation measurements need to be taken to output the instantaneous power generation to an LCD screen, sensors that were rated to handle a high power input needed to be acquired. Some possible sensors looked into were current monitor ICs using high side current shunt and a simple Hall Effect current sensor (uses the electrical fields produced by the current to produce potential). Most current monitor ICs are rated for very low currents therefore they were irrelevant to the application of the project. The Hall Effect sensors seemed promising, but it would have been a lot of extra design work to create the current sensor circuit which would be outside our time constraints. Both the current and voltage ICs were found from a company called Phidgets which provides prebuilt circuits rated high enough to handle our power and still stay within our budget. Once a way to monitor power was discovered, the next roadblock was voltage and current regulation. The Maxim MAX15046 is a 40V, high‐performance, synchronous Buck controller. This chip (while it is complex) provided the exact regulation needed all for less than five dollars per IC. This IC accepts an input voltage from 4.5V to 40V and outputs a fixed voltage that can be configured between the range of .6V to 85% of Vin. It also provides up to a 25A output capability. This will allow stable voltage to be provided to the battery. In parallel with the battery there is a Zener diode to act as overcharge protection. After the battery, the sources run into other regulators that will provide ample voltage to the motors, microprocessor, and other IC chips. Current Monitor Decision Criteria Weights Phidgets 30A Current Monitor Allegro ACS712 TI INA219 Hall Effect Current Monitor Effectiveness 0.4 9 8 3 7 Practicality 0.2 9 5 3 5 Time 0.25 7 2 1 1 Cost 0.15 5 9 7 8 3.1 5.25 TOTALS 1 7.9 6.05 Table 8 ‐ current monitor mechanism decision matrix 20 21
Figure 14 - charge controller circuit diagram
DC‐DC Converter As afore stated, the Maxim MAX15046 is a 40V, high‐performance, synchronous Buck controller is the switching regulator chosen as a part of the charge controller design. This chip (while it is complex) provided the exact regulation needed all for less than five dollars per IC. This IC accepts an input voltage from 4.5V to 40V and outputs a fixed voltage that can be configured between the range of .6V to 85% of Vin. It also provides up to a 25A output capability. This will allow stable voltage to be provided to the battery. The MAX15046 is excellent for the specifications it needs to meet for the project however it requires a lot of external circuitry and can’t be modeled easily is MultiSim. The MAX15046 is a pulse width modulated controller which uses a frequency to step down the voltage from the input. Despite not being able to be easily simulated, the MAX15046 comes with exact specifications how to calculate the values of the external components. Beginning on Appendix A7, the equations for the external components can be found. These equations are given to find the effective input‐voltage range, set the output voltage, set the switching frequency, determine the inductor used, set the valley current limit, and determine the input and output capacitors. 22 Figure 15 - DC-DC step-down voltage
converter circuit diagram
23
Charge Controller Input Simulation
30
Vin
V(after_cap)
I(after_cap)
25
Voltage (V) / Current (A)
20
15
10
5
0
0
2
4
6
8
10
Time (s)
12
14
16
18
20
Microprocessor Design The microprocessor needs to be chosen based upon what inputs and outputs it needs to have for the current project. After the kite system conceptualization, ideas of what is needed in a microprocessor are formed. The need to monitor the current and voltage to calculate the power is first understood. There is also need for an input to measure the length of string released as well as a need for four inputs to measure the kite string angles to determine the kite’s relative position in the air. Finally, a tension sensor input is necessary to be able to scale calculations up or down depending on wind speed. All of these inputs require at least 8 A/D Inputs. A/D inputs will also be needed for the three way user mode switch on the user interface. This switch allows the user to let the kite ascend, retract, or sustain. The other portion of the user interface is the LCD screen which will output the length of line released and the instantaneous power generation measured. The LCD screen will require five I/O ports on the microprocessor. The last feature the microprocessor needs is a pulse width modulation system to control the two motors. Another important consideration taken with the microprocessor is time. Development time is a significant factor in building the electrical system. This led to the need for a development board. A development board allows the programmer to easily write software for the microprocessor and upload it without having the hassle of a using a microchip programmer and designing a PCB board for the microprocessor. The only downfall is the expense that arises with development boards. If a single microprocessor was purchased it would cost $4‐6. The development board cost close to $50. This cost was worth the time that would have been lost if a standalone microprocessor was purchased. To view the trade‐off decision matrix for microprocessor selection, see table 8. All of these considerations led to the Microchip PIC18F4550 Development Board from Futerlec.com. This is a development board within budget that meets all the criteria needed for inputs and outputs, and it even allocates some extra features. All of the sensors will have to be checked on a constant basis and be locked in a continuous loop to keep the kite in flight. The only exceptions will be user‐mode changes from the user interface. This switch will trigger interrupts in the microprocessor and alter the flow of the program which will resume once the interrupt is completed. In figure 17, the string length switch is represented as a simple switch. This will be a ticker switch which will click each time the reel spool rotates. In figure 17, there are also four potentiometers used to represent the kite angle sensors. As the resistance of these changes, the input voltage will change telling the microprocessor the angle change of the string. There are two angle sensors for each string, one for vertical angle and one for horizontal angle. Another potentiometer for the tension sensor is used in figure 17. This is a resistor that varies with the pressure placed on it and will be used to find the tension in the kite strings. The voltage and current sensors will be inputs that vary from 0‐5 volts. These sensors will take a reading on the voltage and current generated by the generator motor and convert that reading to a 0‐5 volt signal. The user interface contains a simple three way switch that will allow the user to select between three modes. The LCD screen will utilize six input/output pins of the microprocessor. The motor controls will utilize six input/output pins as well. Microprocessor Decision 0.4 MicroChip STARTER DEMO BOARD KIT 9 Futurlec MicroChip Development Board 9 Individual MicroChip PIC18F4550 8 Practicality 0.2 8 9 7 Time 0.25 8 9 1 Cost 0.15 1 3 9 TOTALS 1 7.35 8.1 6.2 Criteria Weights Effectiveness Table 8 ‐ microprocessor selection matrix 25 Figure 17 - microprocessor circuit diagram
26
Software Design The actual flow of the software is currently becoming more detailed in development. The kite flight test data is still currently being turned into a mathematical model that can easily be manipulated for software applications. All of the sensors will have to be checked on a constant basis and be locked in a continuous loop to keep the kite in flight. The only exceptions will be user‐mode changes from the user interface. This switch will trigger interrupts in the microprocessor and alter the flow of the program which will resume once the interrupt is completed. The software system will begin by reading the input A/D ports from the voltage and current sensors to monitor the power generated. It will then calculate the power with an equation programmed into memory. This number will then be stored to memory. The program will then reading the input switch sensor for the kite string and line tension using A/D inputs. These values will then be stored to memory as well. The kite string length and generate power will then be taken out of memory and be translated to allow it to be exported to the LCD screen. The export to the screen will then take place. Once the LCD screen has been updated, the software will begin the check kite algorithm. The software will first read in the values of the kite angle potentiometers and calculate the kite string angles. These angles will then be compared with the kite flight algorithms to determine the next move. These movements will then be applied to the control motors to control the kite. The user three way switch is then checked to see if the kite needs to be retracted or ascended. If it needs to change, an output to the spool motor will be applied. This whole process is then repeated. 27 Software Flowchart
START
Find Kite Angles
Get current /
voltage values
Calculate Motor
Movement
Calculate power
P=V*i
Move Control
Motor
Get Kite String
Length and Line
Tension
Check Ascend/
Sustain/Retract
Switch
Output Power and
Kite String Length
to LCD
No
Mode Change?
Yes
Check Kite
Calculate Spool
Motor Movement
Recalibrate Kite
Move Spool Motor
Figure 18 – basic system flowchart
28
Software Flowchart cont’d
START
Find Kite Angles
Get current /
voltage values
Calculate Motor
Movement
Calculate power
P=V*i
Move Control
Motor
Get Kite String
Length and Line
Tension
Check Ascend/
Sustain/Retract
Switch
Output Power and
Kite String Length
to LCD
No
Mode Change?
Yes
Check Kite
Calculate Spool
Motor Movement
Recalibrate Kite
Move Spool Motor
Figure 19 – detailed system flowchart
29
Motor Controller Choosing a motor controller configuration is highly dependent upon the types of motors being used. Two different motors need to be chosen for the current project. One motor, the reel motor, will control the spool that allows the kite to ascend, sustain, or retract. The other motor will adjust the kite controls according to the kite control algorithms. The motor chosen for the reel is a typical 12V DC motor, and the control motor is a 1.76V 2A stepper motor that changes at .8 per degree change. The reel motor will be controlled by an Infineon TLE5205‐2 motor controller chip. This chip is a 5A output H‐bridge made for DC motor management applications. This chip will accept PWM signals from the microprocessor and will turn the motors on and off based upon the inputs received which are set in the programming of the microprocessor. The motors will not need to vary in speed, only in direction, so only the duration the motor is on will need to be adjusted by the software of the microprocessor. Table 9 – function truth table of TLE205‐2 The control motor can be controller with inputs and outputs from the microprocessor. A set of binary codes are programmed and repeatedly sent to the motor to allow it to move to exact specification. A stepper motor is good for the current application because it allows the system to know exactly how far the controls are moving. 30 Electrical Components LCD16x2‐ 16 x 2 Character LCD Display – futurlec.com Figure 20 – 16x2 lcd display Features 16 Characters x 2 Lines 5 x 7 Dots with Cursor Built in Controller +5v Power Supply (Also Available for +3V) 1/16 Duty Circle Phidget 30 AMP Current Sensor AC&DC – trossenrobotics.com Figure 21 – 30A current sensor The formula to translate SensorValue into Current is: DC Amps = (SensorValue / 13.2) ‐ 37.8787 AC RMS Amps = SensorValue x 0.04204 Device Specifications Characteristic Value Active Current Consumption 10mA Output Impedance 1K ohms Maximum Measurable AC Current 30A Maximum Measurable DC Current ±30A Maximum Measurable AC Frequency 10kHz Current Conductor Resistance 1.5mΩ Maximum Supply Voltage 5.5VDC Minimum Supply Voltage 4.5VDC Terminal Block Recommended Wire Size 10 ‐ 26 AWG Wire Stripping Length 6‐7mm Total Output Error 1 ±5% Max between ‐40°C to +85°C Total Output Error (Typical) ±1.5% @ 25°C 31 Battery Decision A large portion of the design decision was reliant upon the battery. The system in creation not only needs to generate power but store it. The chemistry of the battery chosen needs to be looked at closely to make sure it can handle the ruggedness of such a high power system. Since it needs to be a battery that can be recharged, the most likely possible battery chemistries for this system include NiMH, lithium‐ion, and lead acid. After analyzing these battery types, a few conclusions can be drawn. NiMH batteries are too small in capacity and are rated at a current too small for the current project. Lithium‐Ion batteries would possibly work for the current project but are extremely expensive and would be way out of the budget. The last choice is lead acid which applies satisfactorily to this project. Lead acid batteries are rugged and can handle large voltages and currents. Lead acid batteries (even those rated into the dozens of amp‐hours) are moderately affordable. The only disadvantage to lead acid batteries is that a deep discharge can cause an extreme memory loss. The battery chosen is a 12V 26Ah sealed lead acid battery and was purchased for around $50.00. Access SLAA1224F Battery is rated at 12Volts, 26Ah rating. The Access SLAA1224F Battery from AtBatt.com deliver power when you need it and where you need it. It has been specially designed to meet the power needs of your Access SLAA1224F and is maintenance free, easy to handle, rugged and economical. It has a characteristic of high discharge rate, wide operating temperature, long service life and deep discharge recover. Amstron 12 volt 26Ah valve regulated sealed lead acid batteries are maintenance free, rugged and economical. Amstron SLA batteries are utilized in a wide variety of applications including electric vehicles, wheelchairs, scooters, UPS backups, computer systems, industrial and medical equipment and more. Delivering power when you need it, the AP‐12260R uses a state of the art, heavy‐duty, calcium‐alloy grid that provides exceptional performance and service life in both float and cyclic applications. • Absorbent Glass Mat (AGM) technology for superior performance • Valve regulated, spill proof construction allows safe operation in any position • High energy density • Approved for transport by air. • UL recognized under file number MH47341 Specifications Chemistry Lead Acid Voltage 12 Capacity 26,000 mAh / 26 Ah Rating 312 Whr Connector R Terminal Length 6.54 inch / 16.61 cm Width 6.89 inch / 17.50 cm Height 4.92 inch / 12.50 cm Color Gray Weight 18.08 lb / 8,200.91 g Warranty 1 Year UPC Code 880487220654 Figure 22 ‐ Amstron 12V / 26Ah Sealed Lead Acid Battery w/ R Terminal
32
Electronically Assisted Design:
Power from the wind is harnessed from the drag force and movement of the kite parallel to the
kite strings through the power generation system. A kite flight pattern that allows the wind to
produce this force and movement, thereby transferring power, is necessary. It is also important that
this power is delivered in a repeatable method. From the equation for the drag force parallel to the
kite strings (equation 1) we see that the force is a function of how much of the kite area is
perpendicular to the wind. This renders the power zone as shown is figure 23b where the area, hence
forces, are greatest. A kite flying repeatedly in and out of the power zone would produce a figure
eight path, such as is shown in figure 23a This path delivers alternating in and out movements that
pull the freewheel to generate electricity and then allow the freewheel to be retracted in preparation
for another generation pull.
Figure 23a: Sample Flight Path in 3D
Figure 23b: Power Zone
To produce this power generating flight path software, sensors, and electromechanical means
will be used. The kite controls sensors will send analog tension sensor and spherical position signals
to the microprocessor. The electromechanical controls physically control the kite flight through
manipulation of the kite string lengths.
FD = ½ ρV2CLAK
Where CD is the drag coefficient equal to about 1.2, AK is the kite area 2m2, ρ is the density of air at
1.2kg/m2, and V is the wind speed.
Controls Sensors Design:
The sensory requirements for autonomous control of the Second Wind system stand on three
legs. First the location of kite must be sensed, second the velocity vector of the kite must be sensed,
and finally a desired location or path of the kite must be predetermined. The effects of wind speed
are discussed below.
The kite's location can be conveniently measured in spherical coordinates using the kite string
length for the radius r and sensors attached to the kite strings to sense angles φ and θ, as depicted in
figure 24, thus giving the kite's location in spherical coordinates.
33
Figure 24: Kite in Spherical Coordinates
Kite velocity vector can be measured by the microprocessor by comparing the change in kite location
over time. The accuracy of this method will be affected by the sample rate of the kite location. A
single cycle of the flight path was measured to be about 5 seconds with wind speeds around 16 kph
(10 mph). If the sample rate is too slow, say 1 Hz, the information is too old to be useful and if the
sample rate is too fast, say 10000 Hz, the information will be no more than noise. A tentative sample
rate of 40 Hz is proposed giving 200 samples in one cycle. This sample rate will be tested and updated
in the future. Lastly the predetermined path shall be a flight pattern's coordinates that are stored in
the microprocessor. In order for this pattern to be useful in variable wind conditions it must be scaled
in real time proportional to wind speed.
Two options for location/velocity sensors were compared using a decision matrix as shown in
Table 25.
Attribute
Weight
Potentiometer Inclinometer
Cost
40.00%
*10
5
Simplicity
40.00%
7
6
Reliability
20.00%
7
10
Total
100.00%
82.00%
64.00%
Table 10: Location/Velocity Sensor Decision Matrix.
*NOTE: Ratings are on a 0 to 10 scale where 0 is least
desirable and 10 is the most.
A test mock up of the potentiometer sensor was made and its applicability was established. The test
platform was constructed using two potentiometers, various lego parts, and two protractors such that
the Θ and Φ angles of a single angle arm could be measured. This test platform is depicted in Figure
25 below.
34
Figure 25 Test platform
Samples of the Φ potentiometer’s resistances at the angles 0,10,20,30 on until 180 degrees
were taken, recorded, and graphed. Similarly, samples of the Θ potentiometer were taken at angles
0,10,20,30 on until 110 degrees were taken, recorded, and graphed. Trend lines for each graph were
made.
Using the equations generated by the trend lines a program was written in LabView to convert
the measured resistance to an angle measuring device in LabView.
The resulting graphs are shown in Figure 26 and Figure 27 below.
Figure 26: Phi vs R
35
Figure 27: Theta vs R
The results of the program written in LabView for Phi are shown in Figure 28 and the results for Theta
are in Figure 29 below.
Figure 28 Results for Theta
Figure 29 Results for Phi
From Figures 26 and 27 we can see that the resistance change with angle is in fact linear. This
fact is confirmed by the angles indicated in the LabView program output seen in Figures 28 and 29.
The angles read using the program were close if not identical to those observed on the test platform.
The pros of this method of measuring the Θ and Φ are that it is relatively inexpensive and easy to
construct. Possible cons of this method are that is it mechanically intensive (could have problems with
dust, rust, etc. There could also be a problem with drift depending on the quality of the
potentiometer used.
This is a very good potential method of measuring the angles of the kit strings provided dust,
rust and drift are considered in the design.
The total cost for this sensor design is estimated to be between $3 and $5 each and falls under the
miscellaneous electronics category in the budget.
36
In order to scale the predetermined flight pattern proportional to wind speed a tension sensor
was designed. Actual wind speed was not sought because the tensions on the line could be used to
represent both the wind speeds and serve in testing for the efficiency of the overall system. The force
sensor selected for our tension sensor, shown in figure 30, will be incorporated into an overall design
as shown in figures 31 and 32.
8" FlexiForce 0-100 lbs. Resistive Force Sensor Kit:
- 0-100 lbs FlexiForce Force Sensor
- Phidget Voltage Divider
- 24" Sensor Cable
- Price: $24.40
Figure 30: Force Sensor Kit
The tension sensor design shown in figure 32 converts the tension on the string to a force on
the FlexiForce sensor.
Figure 31: Tension Sensor Model
The tension on the string can be calculated from the force measured on the FlexiForce sensor
as shown in the following equation.
37
T = (A*Fs)/*(A+B)*sin(λ)+
Where T is the tension on the string and Fs is the measured force on the sensor. The dimensions A, B,
and λ will be considered constant and are to be measured as shown in figure 32.
Figure 32: Tension Sensor Dimensions
A figure eight flight path was chosen because it readily delivered the repetitive power zone
enter and exit pattern. As was stated previously, the path should be scaled with wind speed. A
depiction of such a path is shown in figure 23a.
This path is ideal because of the power zone shown previously in figure 23b. The kite flies into
the red power zone delivering maximum tension, pulling the freewheel and generating electricity.
Next the kite flies into the low tension light blue zone allowing the free wheel to retract, the kite loops
back and the process is repeated.
38
Controls Kite String Manipulator Design:
The kite's flight is manipulated by pulling the right line IN while letting the left line OUT to turn
CW and pulling the left line IN while letting the right line OUT to turn CCW. To accomplish this a slider
controller is designed such that the linear movement of the slider causes the above mentioned
maneuvers. Our designed controller is depicted in figure 33.
Figure 33: Slider Controller Design
The slider controller is depicted from above in figure 34. It can be seen that when theta equals
zero the line length Y can be calculated to be equal to twice that of the slider position X. For example,
as the slider moves distance X to the right, the right line will bet let out distance 2*X and the left line
will be taken in distance 2*X. The converse is true of the slider moving to the left. This effectively
Figure 34: Slider Controller Design From Above
39
turns the kite. This design is particularly advantageous as line tensions increase because the tensions
will, for the most part, cancel each other leaving the majority of the power required to move the
control slider. This is only valid as the angle θ approaches zero so that the distance D between the eye
hole and the side pulley is the same as the diameter of the central pulleys. The length of a standard
kite control bar is 50cm. The most extreme turning angle achievable with this control bar is reached
when the bar is parallel with the kite string. That is to say one side of the bar has been pulled away
from the kite to the the max and the other has been pushed out towards the kite to the max. The
change in line length during this maneuver is exactly 25cm for each side. This tells us that the
maximum change in line length needed to control the kite is 25cm. Therefore the length (L) for our
slider design must have a minimum of 25cm or roughly 9.9in to sufficiently control the kite. The
difference in the forces on the slider is the difference in the line tensions doubled.
To determine the difference in the line tensions during a turn we must understand the
dynamics taking place during the kite's flight. In figure 35 we see a free body diagram of the kite with
the control slider at neutral. The wind is assumed to be parallel to the ground thereby causing drag
force normal to the inside of the kite. We can see that, because of the slightly tiled orientation of the
kite to the wind, the component of the drag force in the y direction must cause acceleration of the kite
in an ark about the z direction. As the kite flies forward we also see a lift force acting due to the
parafoil shape.
Figure 35: Kite FBD With Control Slider at Neutral
In figure 36 we see an example of a right turn. When the slider is moved to the left the right
string is retracted and the left string is given slack. This re-orients the kite to the wind such that the
drag force normal to the kite now has a component along the z axis. The aerodynamic design of the
kite causes the kite to turn about the x axis as is moves along the z axis.
40
If we assume that the drag
force and lift force along the x axis
are at the center of the kite, that the
kite strings are at an equal distance
Rcenter
from the center, and that there is no
movement in the x direction then we
can say from statics that TR and TL are
equal and there is no difference in
line tension.
ΣMy: TL*Rcenter +TR*Rcenter = 0
Therefore we can say that
there will be no holding force
Figure 36: Kite FBD With Right Turn
needed. The work required to make
the line length change can be calculated as the difference of the work done by the slider to pull one
line in and the work done on the slider by the other line going out.
Wtotal = FS*x - TL *x*2
From the free body diagram in figure 37 we can see that
FS - 2*TR - FFriction= Msystem*aslider .
Plugging this into the work equation we get
Wtotal = (2*TR + Ffriction + Msystem*aslider )*x - TL x*2
Assuming TR = TL this can simplify to
Wtotal = (Ffriction + Msystem*aslider )*x
We see that the amount of work done by
the slider will primarily be a function of
friction in the system and the mass and
acceleration of the system. It can be
assumed that any slider purchased will
already be able to move its own weight.
The only other mass in this system is that
of the kite and the strings which are
negligible. In conclusion; special attention
to the frictional losses to be overcome by
the controller must be considered in the
final selection of small parts that the
strings run through such as pulleys and
eye holes.
Figure 37: Slider Controller FBD
41
The current slider design is made of the components shown in figure 38.
Belt Slider Rail
Truck travel: 13 inches
Cost: $45 (including shipping)
Figure 38: Belt Slider Rail
A motor was purchased to run this slider. The motor is rated at 1.76V 2A and turns at 0.8 deg per
revolution. A cog pulley came with the motor that will mate with the belt drive. This motor coasted
$35 with shipping. This slider meets the length requirements of a 9.9 inch travel length.
Kite Retraction System Design:
The purpose of the kite retraction system is to deploy and retract the kite when the user so
desires or to retract the kite when wind speed is either too high for safety or too low for flight >50 kph
<6 kph.
Two designs were considered, one using a motorized big game fishing reel, and the other using
a simple motor and drum set up. A decision matrix was created to help in the decision process and is
shown in table 11.
Attribute
Weight
Fishing Reel
Simple Drum
Cost
50.00%
*6
10
Simplicity
30.00%
7
10
Consistent Line Intake
20.00%
10
7
Total
100.00%
71.00%
94.00%
Table 11: Kite Retraction Decision Matrix.
*NOTE: Ratings are on a 0 to 10 scale where 0 is least
desirable and 10 is the most.
As is shown above the decision was made to go with a simple drum. In the design
specifications the retraction time is set to be no less than 10 minutes. Using equation 3 it has been
42
calculated that the maximum rated tension on the kite during retraction (when the only force on the
kite will be due to lift) is equivalent to 28.13 N rounded up (about 7lb). A spreadsheet, shown in table
12 was made to find the appropriate spool dimensions, rpm, and torque for these requirements.
FL = ½ ρV2CLAK
Where CL is the lift coefficient equal to about 0.15, AK is the kite area 2m2, ρ is the density of air
at 1.2kg/m2, and V is the wind speed.
Table 12: Excel spreadsheet screen shot.
The equations used to calculate the values in table 12 are as listed below:
Rate line in = string length/retraction time
Rate spool spin = Voltage to Motor * 443rpm/V
-> from CIM motor specks
Length per revolution = Rate line in / Rate spool spin
Design Radius of Spool = sqrt(Length per revolution/[pi*12])
Torque Needed = Design Radius of Spool * 7 lb
Amps to Motor = Torque needed / 0.16124 lb*in/A
-> from CIM motor specks
These values were then converted to metric units as well.
The retraction time value was set to five minutes instead of the required ten minutes in order
to design for a factor of safety of two. Using this spreadsheet a motor was found that met the
specified requirements. The specific motor selected is shown in figure 43 and the resulting system
using this motor is shown in figure 44. It can be calculated that at most about 12.77 watt hours is
required to deploy and retract the kite.
Whtotal = [Whdeploy = 6v*2.7A*(2.5/60)hr] + [Whretract = 6v*24.19A*(5/60)hr] = 12.77 Wh
43
This is insignificant because in the event of a case where wind speeds are high enough to produce
such lift forces on the kite the power generated due to those same high winds would offset any losses
due to the retraction of the kite.
FIRST CIM Motor:
- Operating v : 6v - 12v
- Nominal v : 12v
No Load RPM : 5310
No Load A
: 2.7A
Stall Torque : 343.27 oz-in 2424 mN-m
Stall Current : 133A
Kt
: 2.58 oz-in/A 18.2 mN-m/A
Kv
: 443 rpm/V
Efficiency
: 65%
RPM - Peak Eff : 4614
Torque - Peak Eff
: 45 oz-in/A 317.8 mN-m
Current - Peak Eff
: 19.8A
Figure 39: Retraction Motor
The kite reel system will also act as the kite break system. The cost of parts for the spool drum
will be insignificant due to the fact that its parts can be found in a common junk yard or even made
out of PVC pipe. The value of the spool will be recorded in our records but budget will not suffer.
Figure 40: Retraction System
CIM Motor
Spool Drum
radius ≈ 1.7 cm
(2/3 in)
44
Budget Analysis
Currently 33.3% of the budget has been used by spending $283.09. A large number of
purchases will be made within the next week so that majority of the required parts for the
system will be purchased. Vendors have been selected for all the parts of the project.
Some of the more expensive items:




Kite
Generator Motor
Development Board
Battery
These are some of the more important items that were required in order to further the
research and development of the project.
The generator motor was expected to be more expensive, so this was an unexpected
saving. The budget previously listed the use of used bicycle parts for items such as sprockets,
chains, and axles. However, the design now calls for the use of sprockets and chains of
different pitch from a bicycle system. However, due to the low cost of the motor, the sprockets
and chains still fit within the budget of the generation system.
The battery design was changed from one that could hold 500 watt hours of energy to
one that can hold just over 300 watt hours of energy. This design change was done in order to
save on the cost of the battery. The system will still be able to show that it can produce 500
watt hours of energy within a ten hour period.
A development board was purchased instead of individual microprocessors to allow for
ease and speed of programming. The change proved more expensive however, the
microprocessor that comes with the development board has free samples that can be obtained.
Another major change to the budget comes from the controls system. It was
determined that the most accurate and effective way to control the kite would be by using a
system similar to that used on computer numerical control (CNC) machines. These slider
controller mechanisms can be purchased on ebay, however they are expensive, and will
therefore be a larger part of our budget.
The updated budget has allocated $91.82 or 10.80% of the total budget as a buffer in
case of unpredicted expenditures in the future. So, as of now the project is still on budget.
45
Updated Budget
Product
Kite
Development Board
Battery
Generator Motor
Nuts/Bolts/Wood
Sprockets/Chains/Axle
Axle
Springs
Bearings
Circuit Boards
Controls System
Retraction Motor
Sensors
Electrical Components
Miscellaneous
TOTAL
Amount Spent
80% of Budgeted Items
Vendor
Kite Wind Surf
Futurlec
atbatt.com
Monster Scooter Parts
Lowes
Electric Scooter Parts
Lowes
W.B. Jones Spring Co.
VXB.com
4pcb
ebay.com
Trossen Robotics
Trossen Robotics
allelectronics.com
Figure 41
47
Cost
$131.95
$52.90
$50.33
$47.91
$50.00
$104.87
$5.71
$36.00
$28.43
$50.00
$80.00
$32.00
$78.62
$25.00
$76.28
$850.00
$643.01
$618.98
Ordered
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Schedule Analysis Having a retrospective look at our semester has given us essential insight for the remainder of our project’s development. After the completion of our Requirements Specification our project was well under way. Research began to become more intensive and creativity was vital. As we continued to meet deadlines, our first accomplishment was the successful delivery of our Stage/Gate Presentation as well as the completion of our Final System Overview and Project Plan. Once this was completed, deadlines became more reliant upon the timeframes our own group established. We are persisting through the final steps in the design process. Hands‐on models, circuit simulations and schematics, 3‐D Solidworks models and analyses, and overall systems analyses have been completed. Analyses of Solidworks models and circuit diagrams have taken place in order to justify our design choices. There are currently still finalizations being made to a few subsystems but this should not cause many future delays. These finalizations begin with the kite control algorithms. A little behind schedule, the algorithms are still being modeled in a way that is applicable for software development. This should be finished by the end of the semester but can be finished during the first week of Christmas break if needed. A final decision on the control motor also still needs to be made. The limitations in variety, voltage/current necessities, and funds are making this a difficult search, but it will be decided and ordered by the end of the fall semester. This decision is slightly delaying the design for the motor controller since motor specifications for the control motor are not yet available, only assumptions can be made. Thankfully, the motor controller configuration is simplistic and will be quick and easy to implement (less than a week) once a motor decision is made. This final motor controller design implementation may need to take place over Christmas break, but the spring semester will not be delayed. More research is also being done in the use of the microprocessor in order to begin programming. The microprocessor wasn’t received until December 8, 2009, so strides toward being able to program are well under way. More microprocessor self‐education and programming will take place over Christmas break and will allow the spring semester to run smoothly from a software standpoint. Any delays are only minor design decisions that can be decided before the spring semester. Parts selection is still taking place, however all parts will be ordered before Christmas break and are expected to arrive by January 11. Final analyses and simulations are being completed for use in the final presentation. 48 Gantt Chart - Fall 2009
Second Wind
Josh Dowler, Caleb Meeks, John Snyder
ID
Duration
(Weeks) 8
Finish Date
9/8/2009
9/8/2009
9/29/2009
9/29/2009
10/27/2009
10/13/2009
10/20/2009
9/29/2009
10/20/2009
10/13/2009
9/29/2009
9/29/2009
10/6/2009
10/13/2009
11/3/2009
11/3/2009
9/26/2009
11/17/2009
9/29/2009
10/13/2009
11/17/2009
10/20/2009
11/17/2009
10/27/2009
10/31/2009
11/10/2009
11/10/2009
10/27/2009
10/20/2009
10/13/2009
10/27/2009
11/3/2009
11/14/2009
11/10/2009
11/10/2009
12/7/2009
3
4
7
3
3
2
1.5
5
3
2
3
2
3
3
1.7
1
6.5
1.7
F12.0 System Design / Project Plan
10/1/2009 10/13/2009
1.4
F13.0 Final Design
A1.0 Documentation
A2.0 Project Management
11/17/2009 12/8/2009
9/8/2009 12/10/2009
9/8/2009 12/10/2009
1.8
12.5
12.5
Requirements Specifications
System Overview
Controls Design
Mechanically Governed System Design
Electronically Assissted Design
Kite Retraction System Design
Brake System Design
Generator Design
Freewheel and Return Design
Gearing Ratio Design
Generator Motor Selection
Charge Controller Design
Motor Controller Design
Microprocessor Interface Design
User Interface Configuration Design
System Frame Design
Parts Selection
System Analysis
Sep. 2009
15 22 29
6
Oct. 2009
13 20 27
3
Nov. 2009
10 17 24
1
Dec. 2009
8
Thanksgiving Break
Figure 42
48
Start Date
F1.0
F2.0
F3.0
F3.1
F3.2
F3.3
F3.4
F4.0
F4.1
F4.2
F4.3
F5.0
F6.0
F7.0
F8.0
F9.0
F10.0
F11.0
Task Name
◊
◊
Gantt Chart - Spring 2010
Second Wind
Josh Dowler, Caleb Meeks, John Snyder
ID
Parts Assembly / Testing
Mechanical Control System
Electrical Control System
Kite Reel System
Brake System
Freewheel and Return Mechanism
Gearing Ratio
Generator Motor
Boarding Etching
Charge Controller
Motor Controller
Microprocessor Interface Setup
User Interface Configuration
System Frame Assembly
Programming
Project Status
System Integration
System Testing
Finalize Prototype
Final Project
Documentation
Project Management
Start Date Finish Date
1/11/2010
1/11/2010
1/26/2010
2/9/2010
2/23/2010
2/16/2010
2/2/2010
1/11/2010
1/11/2010
1/26/2010
1/19/2010
2/2/2010
2/23/2010
3/15/2010
1/19/2010
2/16/2010
3/15/2010
4/6/2010
4/13/2010
4/13/2010
1/11/2010
1/11/2010
3/4/2010
2/9/2010
2/16/2010
3/2/2010
3/4/2010
3/4/2010
2/16/2010
2/2/2010
1/21/2010
2/16/2010
2/9/2010
2/23/2010
3/4/2010
4/6/2010
3/4/2010
3/2/2010
4/6/2010
4/26/2010
4/26/2010
4/27/2010
4/29/2010
4/29/2010
Jan. 2010
Duration
(Weeks) 11 19 26
7.7
4.1
3
3
1.6
2.6
2
3.1
1.3
3
3
3
1.6
2
6.6
2
3
2.9
1.9
2
15.3
15.3
2
Feb. 2010
9
16 23
2
9
Mar. 2010
16 23
30
6
13
Apr. 2010
20 27
Spring Break
Figure 43
49
S1.0
S1.1
S1.2
S1.3
S1.4
S1.5
S1.6
S1.7
S1.8
S1.9
S1.10
S1.11
S1.12
S2.0
S3.0
S4.0
S5.0
S6.0
S7.0
S8.0
A1.0
A2.0
Task Name
◊
◊
Appendices
50
Appendix A
MAX15046 DC-DC Converter
19-4719; Rev 0; 7/09
40V, High-Performance, Synchronous
Buck Controller
Features
S Input Voltage Ranges from 4.5V to 40V or 5V
Q10%
S Adjustable Outputs from 0.85 x VIN Down to 0.6V
S Adjustable Switching Frequency (100kHz to 1MHz)
with Q10% (1MHz) Accuracy
S Adaptive Internal Digital Soft-Start
S Up to 25A Output Capability
S Cycle-by-Cycle Valley-Mode Current Limit with
The MAX15046 offers the ability to adjust the switching
frequency from 100kHz to 1MHz with an external resistor. The MAX15046’s adaptive synchronous rectification eliminates the need for an external freewheeling
Schottky diode. The device also utilizes the external
low-side MOSFET’s on-resistance as a current-sense
element, eliminating the need for a current-sense resistor. This protects the DC-DC components from damage
during output overloaded conditions or output shortcircuit faults without requiring a current-sense resistor.
Hiccup-mode current limit reduces power dissipation
during short-circuit conditions. The MAX15046 includes
a power-good output and an enable input with precise
turn-on/turn-off threshold, which can be used for input
supply monitoring and for power sequencing.
S
S
S
S
S
S
S
Adjustable, Temperature-Compensated Threshold
(30mV to 300mV)
Monotonic Startup into Prebiased Output
Q1% Accurate Voltage Reference
3A-Peak Gate Drivers
Hiccup-Mode Short-Circuit Protection (PatentPending Architecture)
Overtemperature Shutdown
Power-Good (PGOOD) Output and Enable Input
(EN) with Q5% Accurate Threshold
Thermally Enhanced 16-Pin QSOP Package
Applications
Industrial Power Supplies (PLC, Industrial
Computers, Fieldbus Components, Fieldbus
Couplers)
Telecom Power Supplies
Base Stations
Additional protection features include sink-mode current
limit, and thermal shutdown. Sink-mode current limit prevents reverse inductor current from reaching dangerous
levels when the device is sinking current from the output.
Ordering Information
PART
The MAX15046 is available in a 16-pin QSOP or 16-pin
QSOP-EP package and operates over the -40NC to +125NC
temperature range.
MAX15046AAEE+
TEMP RANGE
PIN-PACKAGE
-40C to +125C
16 QSOP
MAX15046BAEE+
-40C to +125C 16 QSOP-EP*
+Denotes a lead(Pb)-free/RoHS-compliant package.
*EP = Exposed pad.
Pin Configurations appear at end of data sheet.
Typical Operating Circuit
4.5V TO 40V
VIN
C1
CSP
IN
VCC
MAX15046
PGOOD
ON
Q1
DH
C2
OFF
0.6V TO 0.85V x VIN
VOUT
L1
LX
EN
BST
LIM
DL
Q2
C3
D1
R4
COMP
C5
R3
DRV
C4
C6
C7
R1
FB
PGND
RT
GND
R5
R3
R2
_______________________________________________________________ Maxim Integrated Products 1
For pricing, delivery, and ordering information, please contact Maxim Direct at 1-888-629-4642,
or visit Maxim’s website at www.maxim-ic.com.
MAX15046
General Description
The MAX15046 synchronous step-down controller operates from a 4.5V to 40V input voltage range and generates an adjustable output voltage from 85% of the input
voltage down to 0.6V, supporting loads up to 25A. The
device allows monotonic startup into a prebiased bus
without discharging the output and features adaptive
internal digital soft-start.
MAX15046
40V, High-Performance, Synchronous
Buck Controller
ABSOLUTE MAXIMUM RATINGS
IN to GND ..............................................................-0.3V to +45V
VCC to GND..................... -0.3V to lower of (VIN + 0.6V) and 6V
EN, DRV to GND .....................................................-0.3V to +6V
PGOOD to GND ....................................................-0.3V to +45V
PGND to GND ......................................................-0.3V to +0.3V
DL to PGND.............................................-0.3V to (VDRV + 0.3V)
BST to PGND .......................................................-0.3V to +50V
LX and CSP to PGND...............................................-1V to +45V
LX and CSP to PGND............................-2V (50ns max) to +45V
BST to LX.................................................................-0.3V to +6V
CSP to LX .............................................................-0.3V to +0.3V
DH to LX .................................................. -0.3V to (VBST + 0.3V)
All Other Pins to GND .............................. -0.3V to (VCC + 0.3V)
VCC Short Circuit to GND..........................................Continuous
PGOOD Maximum Sink Current .........................................20mA
Continuous Power Dissipation (TA = +70NC):
16-Pin QSOP (derate 9.6mW/NC above +70NC) .......771.5mW
16-Pin QSOP-EP (derate 22.7mW/NC above +70NC) 1818.2mW
Junction-to-Case Thermal Resistance (θJC) (Note 1)
16-Pin QSOP ................................................................37NC/W
16-Pin QSOP-EP ............................................................6NC/W
Junction-to-Ambient Thermal Resistance (θJA) (Note 1)
16-Pin QSOP ...........................................................103.7NC/W
16-Pin QSOP-EP ..........................................................44NC/W
Operating Temperature Range ........................ -40NC to +125NC
Junction Temperature .....................................................+150NC
Storage Temperature Range............................ -65NC to +150NC
Lead Temperature (soldering, 10s) ................................+300NC
Note 1: Package thermal resistances were obtained using the method described in JEDEC specification JESD51-7, using a four-layer board. For detailed information on package thermal considerations, refer to http://www.maxim-ic.com/thermal-tutorial.
Stresses beyond those listed under “Absolute Maximum Ratings” may cause permanent damage to the device. These are stress ratings only, and functional operation of the device at these
or any other conditions beyond those indicated in the operational sections of the specifications is not implied. Exposure to absolute maximum rating conditions for extended periods may
affect device reliability.
ELECTRICAL CHARACTERISTICS
(VIN = 24V, VEN = 5V, VGND = VPGND = 0V, CIN = 1FF, CVCC = 4.7FF, RRT = 49.9kI, TA = TJ = -40NC to +125NC, unless otherwise
noted. Typical values are at TA = +25NC.) (Note 2)
PARAMETER
SYMBOL
CONDITIONS
MIN
TYP
MAX
UNITS
SYSTEM SPECIFICATIONS
Input-Voltage Range
VIN
Quiescent Supply Current
IIN_Q
Shutdown Supply Current
IIN_SBY
VIN = VCC = VDRV
4.5
40
4.5
5.5
VIN = 24V, VFB = 0.9V, no switching
VIN = 24V, VEN = 0V, IVCC = 0,
PGOOD = unconnected
V
2
3
mA
0.35
0.55
mA
5.25
5.5
V
0.18
0.45
V
VCC REGULATOR
Output Voltage
VVCC
6V ≤ VIN ≤ 40V, ILOAD = 6mA
5
VCC Regulator Dropout
VIN = 4.5V, ILOAD = 25mA
VCC Short-Circuit Output Current
VIN = 5V
30
55
90
mA
VVCC rising
3.8
4
4.2
V
VCC Undervoltage Lockout
VCCUVLO
VCC Undervoltage Lockout
Hysteresis
400
mV
ERROR AMPLIFIER (FB, COMP)
FB Input-Voltage Set Point
VFB
FB Input Bias Current
IFB
VFB = 0.6V
-250
FB to COMP Transconductance
gM
ICOMP = Q20FA
600
584
Open-Loop Gain
Unity-Gain Bandwidth
2
Capacitor from COMP to GND =
47pF
590
1200
596
mV
+250
nA
1800
FS
80
dB
5
MHz
______________________________________________________________________________________
40V, High-Performance, Synchronous
Buck Controller
(VIN = 24V, VEN = 5V, VGND = VPGND = 0V, CIN = 1FF, CVCC = 4.7FF, RRT = 49.9kI, TA = TJ = -40NC to +125NC, unless otherwise
noted. Typical values are at TA = +25NC.) (Note 2)
PARAMETER
SYMBOL
CONDITIONS
MIN
VCOMP-RAMP Minimum Voltage
COMP Source/Sink Current
TYP
MAX
200
ICOMP
VCOMP = 1.4V
EN Input High
VEN_H
VEN rising
EN Input Low
VEN_L
VEN falling
IEN
VEN = 5.5V
50
UNITS
mV
80
110
FA
1.20
1.26
V
ENABLE (EN)
EN Input Leakage Current
1.14
1.05
-1
V
+1
FA
kHz
OSCILLATOR
Switching Frequency (100kHz)
fSW
RRT = 150kI
80
100
120
Switching Frequency (300kHz)
fSW
RRT = 49.9kI
270
300
330
kHz
Switching Frequency (1MHz)
fSW
RRT = 14.3kI
0.9
1
1.1
MHz
(Note 3)
100
1000
kHz
RRT = 49.9kI
1.15
1.25
V
Switching Frequency Adjustment
Range
RT Voltage
VRT
1.2
PWM MODULATOR
PWM Ramp Peak-to-Peak
Amplitude
PWM Ramp Valley
VRAMP
1.5
VVALLEY
1.5
Minimum Controllable On-Time
70
Maximum Duty Cycle
fSW = 300kHz (RRT = 49.9kI)
Minimum Low-Side On-Time
fSW = 1MHz (RRT = 49.9kI)
85
V
V
125
ns
87.5
%
110
ns
OUTPUT DRIVERS/DRIVERS SUPPLY (VDRV)
Undervoltage Lockout
VDRV_UVLO
VDRV rising
4.0
DRV Undervoltage Lockout
Hysteresis
DH On-Resistance
DL On-Resistance
DH Peak Current
4.2
4.4
400
Low, sinking 100mA,
VBST - VLX = 5V
High, sourcing 100mA,
VBST - VLX = 5V
Low, sinking 100mA,
VDRV = VCC = 5.25V
High, sourcing 100mA,
VDRV = VCC = 5.25V
CLOAD = 10nF
V
mV
1
3
1.5
4
I
1
3
1.5
4
Sinking,
VBST - VLX = 5V
3
Sourcing,
VBST - VLX = 5V
2
A
_______________________________________________________________________________________
3
MAX15046
ELECTRICAL CHARACTERISTICS (continued)
MAX15046
40V, High-Performance Synchronous
Buck Controller
ELECTRICAL CHARACTERISTICS (continued)
(VIN = 24V, VEN = 5V, VGND = VPGND = 0V, CIN = 1FF, CVCC = 4.7FF, RRT = 49.9kI, TA = TJ = -40NC to +125NC, unless otherwise
noted. Typical values are at TA = +25NC.) (Note 2)
PARAMETER
SYMBOL
CONDITIONS
CLOAD = 10nF
DL Peak Current
MIN
TYP
Sinking, VDRV = VCC
= 5.25V
3
Sourcing, VDRV =
VCC = 5.25V
2
MAX
UNITS
A
DH, DL Break-Before-Make Time
(Dead Time)
10
ns
2048
Switching
Cycles
64
Steps
SOFT-START
Soft-Start Duration
Reference Voltage Steps
CURRENT LIMIT/HICCUP
Cycle-by-Cycle Valley CurrentLimit Threshold Adjustment
Range
LIM Reference Current
VCSP - VPGND,
valley limit =
VLIM/10
ILIM
VLIM = 0.3V
30
VLIM = 3V
300
mV
VLIM = 0.3V to 3V, TA = +25NC
45
LIM Reference Current
Temperature Coefficient
50
55
2300
CSP Input Bias Current
VCSP = 40V
-1
Number of Consecutive CurrentLimit Events to Hiccup
Hiccup Timeout
VCSP - VPGND, sink limit = VLIM/20,
RILIM = 30kI, VLIM = 1.5V,
TA = +25NC
Peak Low-Side Sink Current-Limit
Threshold
FA
ppm/NC
+1
FA
7
Events
4096
Switching
Cycles
75
mV
POWER-GOOD (PGOOD)
PGOOD Threshold
VFB rising
PGOOD Threshold Hysteresis
VFB falling
PGOOD Output Low Voltage
VPGOOD_L
PGOOD Output Leakage Current
ILEAK_PGOOD
90
94
IPGOOD = 2mA, VEN = 0V
VPGOOD = 40V, VEN = 5V, VFB = 1V
97.5
2.65
-1
%VFB
%VFB
0.4
V
+1
FA
THERMAL SHUTDOWN
Thermal Shutdown Threshold
Temperature rising
Thermal Shutdown Hysteresis
+150
NC
20
NC
Note 2: All devices are 100% tested at room temperature and guaranteed by design over the specified temperature range.
Note 3: Select RRT as: RRT 4
17.3 10 9
fSW (1 x 10 -7 )(fSW 2 )
, where fSW is in Hertz.
______________________________________________________________________________________
MAX15046
40V, High-Performance, Synchronous
Buck Controller
Pin Description
PIN
8
NAME
FUNCTION
1
IN
Regulator Input. Connect to the input rail of the buck converter. Bypass IN to PGND with a 100nF
minimum ceramic capacitor. When operating in the 5V Q10% range, connect IN to VCC.
2
VCC
5.25V Linear Regulator Output. Bypass VCC to PGND with a ceramic capacitor of at least 4.7FF
when VCC supplies MOSFET gate-driver current at DRV or 2.2FF when VCC is not used to power
DRV.
3
PGOOD
4
EN
Active-High Enable Input. Pull EN to GND to disable the buck converter output. Connect to VCC
for always-on operation. EN can be used for power sequencing and as a UVLO adjustment input.
5
LIM
Current-Limit Input. Connect a resistor from LIM to GND to program the current-limit threshold from
30mV (RLIM = 6kI) to 300mV (RLIM = 60kI).
6
COMP
Error-Amplifier Output. Connect compensation network from COMP to FB or from COMP to GND.
7
FB
Feedback Input (Inverting Input of Error Amplifier). Connect FB to a resistive divider between the
buck converter output and GND to adjust the output voltage from 0.6V up to 0.85 x IN.
8
RT
Oscillator-Timing Resistor Input. Connect a resistor from RT to GND to set the oscillator frequency
from 100kHz to 1MHz.
Open-Drain Power-Good Output. Pull up PGOOD to an external power supply or output with an
external resistor.
9
GND
10
PGND
Analog Ground. Connect PGND and AGND together at a single point.
11
DRV
12
DL
Low-Side External MOSFET Gate-Driver Output. DL swings from DRV to PGND.
13
BST
Boost Flying Capacitor Connection. Internally connected to the high-side driver supply. Connect a
ceramic capacitor of at least 100nF between BST and LX and a diode between BST and DRV for
the high-side MOSFET gate-driver supply.
14
LX
Inductor Connection. Also serves as a return terminal for the high-side MOSFET driver current.
Connect LX to the switching side of the inductor.
15
DH
High-Side External MOSFET Gate-Driver Output. DH swings from BST to LX.
16
CSP
Current-Sense Positive Input. Connect to the drain of low-side MOSFET with Kelvin connection.
—
EP
Power Ground. Use PGND as a return path for the low-side MOSFET gate driver.
Gate-Driver Supply Voltage. DRV is internally connected to the low-side driver supply. Bypass DRV to
PGND with a 2.2FF minimum ceramic capacitor (see the Typical Application Circuits).
Exposed Pad. EP is internally connected to ground. Connect EP to a large copper ground plane to
maximize thermal performance.
______________________________________________________________________________________
MAX15046
40V, High-Performance, Synchronous
Buck Controller
Detailed Description
The MAX15046 synchronous step-down controller operates from a 4.5V to 40V input-voltage range and generates an adjustable output voltage from 85% of the inputvoltage down to 0.6V while supporting loads up to 25A.
As long as the device supply voltage is within 5.0V to
5.5V, the input power bus (VIN) can be as low as 3.3V.
The MAX15046 offers adjustable switching frequency
from 100kHz to 1MHz with an external resistor. The
adjustable switching frequency provides design flexibility in selecting passive components. The MAX15046
adopts an adaptive synchronous rectification to eliminate external freewheeling Schottky diodes and improve
efficiency. The device utilizes the on-resistance of the
external low-side MOSFET as a current-sense element.
The current-limit threshold voltage is resistor-adjustable
from 30mV to 300mV and is temperature-compensated,
so that the effects of the MOSFET RDS(ON) variation
over temperature are reduced. This current-sensing
scheme protects the external components from damage
during output overloaded conditions or output shortcircuit faults without requiring a current-sense resistor.
Hiccup-mode current limit reduces power dissipation
during short-circuit conditions. The MAX15046 includes
a power-good output and an enable input with precise
turn-on/-off threshold to be used for monitoring and for
power sequencing.
The MAX15046 features internal digital soft-start that
allows prebias startup without discharging the output. The
digital soft-start function employs sink current limiting to
prevent the regulator from sinking excessive current when
the prebias voltage exceeds the programmed steadystate regulation level. The digital soft-start feature prevents
the synchronous rectifier MOSFET and the body diode of
the high-side MOSFET from experiencing dangerous levels of current while the regulator is sinking current from the
output. The MAX15046 shuts down at a +150NC junction
temperature to prevent damage to the device.
DC-DC PWM Controller
The MAX15046 step-down controller uses a PWM voltage-mode control scheme (see the Functional Diagram).
Control-loop compensation is external for providing maximum flexibility in choosing the operating frequency and
output LC filter components. An internal transconductance error amplifier produces an integrated error voltage at COMP that helps to provide higher DC accuracy.
The voltage at COMP sets the duty cycle using a PWM
10
comparator and a ramp generator. On the rising edge
of an internal clock, the high-side n-channel MOSFET
turns on and remains on until either the appropriate duty
cycle or the maximum duty cycle is reached. During
the on-time of the high-side MOSFET, the inductor current ramps up. During the second-half of the switching
cycle, the high-side MOSFET turns off and the low-side
n-channel MOSFET turns on. The inductor releases the
stored energy as the inductor current ramps down, providing current to the output. Under overload conditions,
when the inductor current exceeds the selected valley
current-limit threshold (see the Current-Limit Circuit (LIM)
section), the high-side MOSFET does not turn on at the
subsequent clock rising edge and the low-side MOSFET
remains on to let the inductor current ramp down.
Internal 5.25V Linear Regulator
An internal linear regulator (VCC) provides a 5.25V nominal supply to power the internal functions and to drive the
low-side MOSFET. Connect IN and VCC together when
using an external 5V Q10% power supply. The maximum
regulator input voltage (VIN) is 40V. Bypass IN to GND
with a 1FF ceramic capacitor. Bypass the output of the
linear regulator (VCC) with a 4.7FF ceramic capacitor to
GND. The VCC dropout voltage is typically 180mV. When
VIN is higher than 5.5V, VVCC is typically 5.25V. The
MAX15046 also employs an undervoltage lockout circuit
that disables the internal linear regulator when VVCC
falls below 3.6V (typical). The 400mV UVLO hysteresis
prevents chattering on power-up/power-down.
MOSFET Gate Drivers (DH, DL)
DH and DL are optimized for driving large-size n-channel
power MOSFETs. Under normal operating conditions and
after startup, the DL low-side drive waveform is always
the complement of the DH high-side drive waveform,
with controlled dead time to prevent crossconduction or
“shoot-through.” An adaptive dead-time circuit monitors
the DH and DL outputs and prevents the opposite-side
MOSFET from turning on until the MOSFET is fully off.
Thus, the circuit allows the high-side driver to turn on
only when the DL gate driver has turned off preventing
the low side (DL) from turning on until the DH gate driver
has turned off.
The adaptive driver dead time allows operation without
shoot-through with a wide range of MOSFETs, minimizing delays and maintaining efficiency. There must be a
low-resistance, low-inductance path from DL and DH to
the MOSFET gates for the adaptive dead-time circuits
_____________________________________________________________________________________
40V, High-Performance, Synchronous
Buck Controller
Undervoltage Lockout
The MAX15046 provides an internal undervoltage lockout (UVLO) circuit to monitor the voltage on VCC. The
UVLO circuit prevents the MAX15046 from operating
when VCC is lower than VUVLO. The UVLO threshold is
4V, with 400mV hysteresis to prevent chattering on the
rising/falling edge of the supply voltage. DL and DH stay
low to inhibit switching when the device is in undervoltage lockout.
Thermal-Overload Protection
Thermal-overload protection limits total power dissipation in the MAX15046. When the junction temperature of
the device exceeds +150NC, an on-chip thermal sensor
shuts down the device, forcing DL and DH low, which
allows the device to cool. The thermal sensor turns the
device on again after the junction temperature cools by
20NC. The regulator shuts down and soft-start resets
during thermal shutdown. Power dissipation in the LDO
regulator and excessive driving losses at DH/DL trigger
thermal-overload protection. Carefully evaluate the total
power dissipation (see the Power Dissipation section) to
avoid unwanted triggering of the thermal-overload protection in normal operation.
Applications Information
The maximum voltage conversion ratio is limited by the
maximum duty cycle (Dmax):
VOUT
D
VDROP2 (1-D max ) VDROP1
D max - max
VIN
VIN
where VDROP1 is the sum of the parasitic voltage drops
in the inductor discharge path, including synchronous
rectifier, inductor, and PCB resistance. VDROP2 is the
sum of the resistance in the charging path, including
high-side switch, inductor, and PCB resistance. In practice, provide adequate margin to the above conditions
for good load-transient response.
Setting the Output Voltage
Set the MAX15046 output voltage by connecting a resistive divider from the output to FB to GND (Figure 2).
Select R2 from between 4kI and 16kI. Calculate R1
with the following equation:
V
R1 R 2 OUT -1
V
FB where VFB = 0.59V (see the Electrical Characteristics
table) and VOUT can range from 0.6V to (0.85 O VIN).
Resistor R1 also plays a role in the design of the Type
III compensation network. Review the values of R1 and
R2 when using a Type III compensation network (see the
Type III Compensation Network (Figure 4) section).
Effective Input-Voltage Range
The MAX15046 operates from 4.5V to 40V input supplies
and regulates output down to 0.6V. The minimum voltage
conversion ratio (VOUT/VIN) is limited by the minimum
controllable on-time. For proper fixed-frequency PWM
operation, the voltage conversion ratio must obey the
following condition:
VOUT
t ON(MIN) fSW
VIN
where tON(MIN) is 125ns and fSW is the switching frequency in Hertz. Pulse skipping occurs to decrease the
effective duty cycle when the desired voltage conversion
does not meet the above condition. Decrease the switching frequency or lower the input voltage VIN to avoid
pulse skipping.
OUT
R1
FB
MAX15046
R2
Figure 2. Adjustable Output Voltage
______________________________________________________________________________________
13
MAX15046
MAX15046 stops both DL and DH drivers and waits for
4096 switching cycles (hiccup timeout delay) before
attempting a new soft-start sequence. The hiccup-mode
protection remains active during the soft-start time.
MAX15046
40V, High-Performance, Synchronous
Buck Controller
Setting the Switching Frequency
An external resistor connecting RT to GND sets the
switching frequency (fSW). The relationship between fSW
and RRT is:
17.3 10 9
R RT fSW (1x10 -7 ) x (fSW 2 )
where fSW is in kHz and RRT is in kI. For example, a
300kHz switching frequency is set with RRT = 49.9kI.
Higher frequencies allow designs with lower inductor
values and less output capacitance. Peak currents and
I2R losses are lower at higher switching frequencies, but
core losses, gate-charge currents, and switching losses
increase.
Inductor Selection
Three key inductor parameters must be specified for
operation with the MAX15046: inductance value (L),
inductor saturation current (ISAT), and DC resistance
(RDC). To determine the inductance, select the ratio of
inductor peak-to-peak AC current to DC average current (LIR) first. For LIR values that are too high, the RMS
currents are high, and therefore I2R losses are high.
Use high-valued inductors to achieve low LIR values.
Typically, inductor resistance is proportional to inductance for a given package type, which again makes I2R
losses high for very low LIR values. A good compromise
between size and loss is a 30% peak-to-peak ripple current to average-current ratio (LIR = 0.3). The switching
frequency, input voltage, output voltage, and selected
LIR determine the inductor value as follows:
L
VOUT (VIN - VOUT )
VIN fSW I OUT LIR
where VIN, VOUT, and IOUT are typical values. The
switching frequency is set by RT (see Setting the
Switching Frequency section). The exact inductor value
is not critical and can be adjusted to make trade-offs
among size, cost, and efficiency. Lower inductor values minimize size and cost, but also improve transient
response and reduce efficiency due to higher peak currents. On the other hand, higher inductance increases
efficiency by reducing the RMS current.
saturation current rating (ISAT) must be high enough to
ensure that saturation cannot occur below the maximum
current-limit value (ICL(MAX)), given the tolerance of the
on-resistance of the low-side MOSFET and of the LIM
reference current (ILIM). Combining these conditions,
select an inductor with a saturation current (ISAT) of:
I SAT 1.35 I CL(TYP)
where ICL(TYP) is the typical current-limit set point. The
factor 1.35 includes RDS(ON) variation of 25% and 10%
for the LIM reference current error. A variety of inductors
from different manufacturers are available to meet this
requirement (for example, Vishay IHLP-4040DZ-1-5 and
other inductors from the same series).
Setting the Valley Current Limit
The minimum current-limit threshold must be high enough
to support the maximum expected load current with the
worst-case low-side MOSFET on-resistance value as the
RDS(ON) of the low-side MOSFET is used as the currentsense element. The inductor’s valley current occurs at
ILOAD(MAX) minus one half of the ripple current. The
minimum value of the current-limit threshold voltage
(VITH) must be higher than the voltage on the low-side
MOSFET during the ripple-current valley,
LIR VITH RDS(ON,MAX) ILOAD(MAX) 1 2 where RDS(ON,MAX) in I is the maximum on-resistance
of the low-side MOSFET at maximum load current
ILOAD(MAX) and is calculated from the following equation:
R DS(ON,MAX) R DS(ON) [1 TC MOSFET (TMAX - TAMB )]
where RDS(ON) (in I is the on-resistance of the lowside MOSFET at ambient temperature TAMB (in degrees
Celsius), TCMOSFET is the temperature coefficient of
the low-side MOSFET in ppm/NC, and TMAX (in degrees
Celsius) is the temperature at maximum load current
ILOAD(MAX). Obtain the RDS(ON) and TCMOSFET from the
MOSFET data sheet.
Find a low-loss inductor with the lowest possible DC
resistance that fits in the allotted dimensions. The
14
_____________________________________________________________________________________
40V, High-Performance, Synchronous
Buck Controller
R LIM 50 10
6
10 VITH
1 2300 (TMAX - TAMB) where RLIM is in I, VITH is in V, TMAX and TAMB are in
NC.
An RLIM resistance range of 6kI to 60kI corresponds
to a current-limit threshold of 30mV to 300mV. Use 1%
tolerance resistors when adjusting the current limit to
minimize error in the current-limit threshold.
Input Capacitor
The input filter capacitor reduces peak current drawn
from the power source and reduces noise and voltage
ripple on the input caused by the switching circuitry. The
input capacitor must meet the ripple current requirement
(IRMS) imposed by the switching currents as defined by
the following equation:
IRMS ILOAD(MAX)
VOUT (VIN - VOUT )
VIN
IRMS attains a maximum value when the input voltage equals twice the output voltage (VIN = 2VOUT),
so IRMS(MAX) = ILOAD(MAX)/2. For most applications,
nontantalum capacitors (ceramic, aluminum, polymer, or
OS-CON) are preferred at the inputs due to the robustness of nontantalum capacitors to accommodate high
inrush currents of systems being powered from very low
impedance sources. Additionally, two (or more) smallervalue low-ESR capacitors should be connected in parallel to reduce high-frequency noise.
Output Capacitor
The key selection parameters for the output capacitor
are capacitance value, ESR, and voltage rating. These
parameters affect the overall stability, output ripple voltage, and transient response. The output ripple has two
components: variations in the charge stored in the output
capacitor, and the voltage drop across the capacitor’s
ESR caused by the current flowing into and out of the
capacitor:
DVRIPPLE = DVESR + DVQ
The output-voltage ripple as a consequence of the ESR
and the output capacitance is:
VESR IP-P ESR
IP-P
VQ 8 C OUT fSW
V - VOUT VOUT IP-P IN
fSW L VIN where IP-P is the peak-to-peak inductor current ripple
(see the Inductor Selection section). Use these equations for initial capacitor selection. Decide on the final
values by testing a prototype or an evaluation circuit.
Check the output capacitor against load-transient
response requirements. The allowable deviation of the
output voltage during fast load transients determines
the capacitor output capacitance, ESR, and equivalent
series inductance (ESL). The output capacitor supplies
the load current during a load step until the controller
responds with a higher duty cycle. The response time
(tRESPONSE) depends on the closed-loop bandwidth of
the converter (see the Compensation Design section).
The resistive drop across the ESR of the output capacitor, the voltage drop across the ESL (DVESL) of the
capacitor, and the capacitor discharge, cause a voltage
droop during the load step.
Use a combination of low-ESR tantalum/aluminum electrolytic and ceramic capacitors for improved transient
load and voltage ripple performance. Nonleaded capacitors and capacitors in parallel help reduce the ESL.
Keep the maximum output-voltage deviation below the
tolerable limits of the load. Use the following equations to
calculate the required ESR, ESL, and capacitance value
during a load step:
ESR VESR
I STEP
I
t
C OUT STEP RESPONSE
VQ
ESL VESL t STEP
I STEP
t RESPONSE 1
3 fO
where ISTEP is the load step, tSTEP is the rise time of the
load step, tRESPONSE is the response time of the controller, and fO is the closed-loop crossover frequency.
______________________________________________________________________________________
15
MAX15046
Connect an external resistor (RLIM) from LIM to GND
to adjust the current-limit threshold, which is temperature-compensated with a temperature coefficient of
-2300ppm/NC. The relationship between the current-limit
threshold (VITH) and RLIM is:
Appendix B
PIC18F4550 Microprocessor
Appendix C
TLE5205-2 Motor Controller
5-A H-Bridge for DC-Motor Applications
1
Overview
1.1
Features
•
•
•
•
•
•
•
•
•
•
•
•
Delivers up to 5 A continuous 6 A peak current
Optimized for DC motor management applications
Operates at supply voltages up to 40 V
Very low RDS ON; typ. 200 mΩ @ 25 °C per switch
Output full short circuit protected
Overtemperature protection with hysteresis
and diagnosis
Short circuit and open load diagnosis
with open drain error flag
Undervoltage lockout
CMOS/TTL compatible inputs with hysteresis
No crossover current
Internal freewheeling diodes
Wide temperature range; − 40 °C < Tj < 150 °C
Type
Ordering Code Package
TLE 5205-2
Q67000-A9283 P-TO220-7-11
TLE 5205-2GP
Q67006-A9237 P-DSO-20-12
TLE 5205-2G
Q67006-A9325 P-TO263-7-1
TLE 5205-2S
Q67000-A9324 P-TO220-7-12
TLE 5205-2
P-TO220-7-11
P-DSO-20-12
P-TO263-7-1
Description
P-TO220-7-12
The TLE 5205-2 is an integrated power H-bridge with
DMOS output stages for driving DC-Motors. The part is
built using the Infineon multi-technology process SPT® which allows bipolar and CMOS
control circuitry plus DMOS power devices to exist on the same monolithic structure.
Operation modes forward (cw), reverse (ccw), brake and high impedance are invoked
from just two control pins with TTL/CMOS compatible levels. The combination of an
extremely low RDS ON and the use of a power IC package with low thermal resistance and
high thermal capacity helps to minimize system power dissipation. A blocking capacitor
at the supply voltage is the only external circuitry due to the integrated freewheeling
diodes.
Data Sheet
1
2001-06-19
TLE 5205-2
Overview
1.2
Pin Configuration (top view)
TLE 5205-2
1
2
3
4
5
TLE 5205-2GP
6
7
GND
N.C.
N.C.
N.C.
N.C.
VS
Q1
EF
IN1
GND
1
2
3
4
5
6
7
8
9
10
20
19
18
17
16
15
14
13
12
11
GND
N.C.
N.C.
N.C.
N.C.
VS
Q2
N.C.
IN2
GND
AEP01680
TLE 5205-2S
EF
OUT1
GND
IN1
VS
IN2
OUT2
AEP01990
TLE 5205-2G
1
2 3
4 5 6
1 2
7
3
4
5
6 7
OUT1 IN1
IN2 OUT2
EF GND V S
AEP01991
OUT1
EF
IN1
IN2
GND
OUT2
VS
AEP02513
Figure 1
Data Sheet
2
2001-06-19
TLE 5205-2
Overview
1.3
Pin Definitions and Functions
Pin No.
P-TO220
Pin No.
P-DSO
Symbol
Function
1
7
OUT1
Output of Channel 1; Short-circuit protected;
integrated freewheeling diodes for inductive loads.
2
8
EF
Error Flag; TTL/CMOS compatible output
for error detection; (open drain)
3
9
IN1
Control Input 1;
TTL/CMOS compatible
4
1, 10,
11, 20
GND
Ground;
internally connected to tab
5
12
IN2
Control Input 2;
TTL/CMOS compatible
6
6, 15
VS
Supply Voltage; block to GND
7
14
OUT2
Output of Channel 2; Short-circuit protected;
integrated freewheeling diodes for inductive loads.
–
2, 3, 4, 5, N.C.
16, 17, 18,
19
Data Sheet
Not Connected
3
2001-06-19
TLE 5205-2
Overview
1.4
Functional Block Diagram
VS
EF
6
2
Error Flag
Diagnosis and Protection Circuit 1
IN1
IN2
IN
3
5
1
OUT
1 2
1
2
0
0
1
1
0 1 0
1 0 1
0 0 0
1 Z Z
7
OUT1
OUT2
Diagnosis and Protection Circuit 2
4
GND
Figure 2
Data Sheet
AEB02394
Block Diagram
4
2001-06-19
TLE 5205-2
Overview
1.5
Circuit Description
Input Circuit
The control inputs consist of TTL/CMOS-compatible schmitt-triggers with hysteresis.
Buffer amplifiers are driven by this stages.
Output Stages
The output stages consist of a DMOS H-bridge. Integrated circuits protect the outputs
against short-circuit to ground and to the supply voltage. Positive and negative voltage
spikes, which occur when switching inductive loads, are limited by integrated
freewheeling diodes.
A monitoring circuit for each output transistor detects whether the particular transitor is
active and in this case prevents the corresponding source transistor (sink transistor) from
conducting in sink operation (source operation). Therefore no crossover currents can
occur.
1.6
Input Logic Truth Table
Functional Truth Table
IN1
IN2
OUT1
OUT2
Comments
L
L
H
L
Motor turns clockwise
L
H
L
H
Motor turns counterclockwise
H
L
L
L
Brake; both low side transistors turned-ON
H
H
Z
Z
Open circuit detection
Notes for Output Stage
Symbol
Value
L
Low side transistor is turned-ON
High side transistor is turned-OFF
H
High side transistor is turned-ON
Low side transistor is turned-OFF
Z
High side transistor is turned-OFF
Low side transistor is turned-OFF
Data Sheet
5
2001-06-19
Was this manual useful for you? yes no
Thank you for your participation!

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

Download PDF

advertisement