Executive Summary - Iowa State University



Iowa State UniversityWind Turbine Energy Conversion System Design and IntegrationProject PlanProject NumberMay10-17Team MembersBrandon JanssenKenny ThelenHassan BurawiElsammani AhmedAdvisorDr. Venkataramana AjjarapuClientISU Department of Electrical and Computer EngineeringDISCLAIMER: This document was developed as a part of the requirements of an Electrical and Computer engineering course at Iowa State University, Ames, Iowa. This document does not constitute a professional engineering design or a professional land surveying document. Although the information is intended to be accurate, the associated students, faculty, and Iowa State University make no claims, promises, or guarantees about the accuracy, completeness, quality, or adequacy of the information. The user of this document shall ensure that any such use does not violate any laws with regard to professional licensing and certification requirements. This use includes any work resulting from this student-prepared document that is required to be under the responsible charge of a licensed engineer or surveyor. This document is copyrighted by the students who produced this document and the associated faculty advisors. No part may be reproduced without the written permission of the Senior Design course coordinator.Contents TOC \o "1-3" \h \z \u Executive Summary PAGEREF _Toc260187452 \h 5Problem Statement PAGEREF _Toc260187453 \h 5Operating Environment PAGEREF _Toc260187454 \h 6Intended Users and Uses PAGEREF _Toc260187455 \h 6Expected End Product and Deliverables PAGEREF _Toc260187456 \h 6Requirements PAGEREF _Toc260187457 \h 7Functional requirements PAGEREF _Toc260187458 \h 7Non-functional requirements PAGEREF _Toc260187459 \h 7Market Survey PAGEREF _Toc260187460 \h 7Resources and Schedule PAGEREF _Toc260187461 \h 8Parts PAGEREF _Toc260187462 \h 8Hours PAGEREF _Toc260187463 \h 8Schedule PAGEREF _Toc260187464 \h 9Design PAGEREF _Toc260187465 \h 10Conceptual Sketch/Update Block Diagram PAGEREF _Toc260187466 \h 10Conceptual Sketch Figure 1 PAGEREF _Toc260187467 \h 10Wind Turbine PAGEREF _Toc260187468 \h 11Inverter PAGEREF _Toc260187469 \h 11Controls PAGEREF _Toc260187470 \h 12Interface PAGEREF _Toc260187471 \h 12DC Sensors PAGEREF _Toc260187472 \h 13RPM Sensor PAGEREF _Toc260187473 \h 13RPM Sensor Circuit Schematic PAGEREF _Toc260187474 \h 14Three Situations of Concern PAGEREF _Toc260187475 \h 16Implementation PAGEREF _Toc260187476 \h 17Interface PAGEREF _Toc260187477 \h 17DC Sensors PAGEREF _Toc260187478 \h 19RPM Sensor PAGEREF _Toc260187479 \h 20Power Supply PAGEREF _Toc260187480 \h 20Three-Phase Induction Motor PAGEREF _Toc260187481 \h 21Press-fit Precision Metal Bellows Couplings: PAGEREF _Toc260187482 \h 21Load PAGEREF _Toc260187483 \h 22Testing PAGEREF _Toc260187484 \h 23Interface PAGEREF _Toc260187485 \h 23DC Sensors PAGEREF _Toc260187486 \h 23RPM Sensor PAGEREF _Toc260187487 \h 23Complete System PAGEREF _Toc260187488 \h 24Power vs. Speed PAGEREF _Toc260187489 \h 24Conclusion PAGEREF _Toc260187490 \h 25Operation Manual PAGEREF _Toc260187491 \h 26Appendix PAGEREF _Toc260187492 \h 29LEM LA 55-P Current Transducer PAGEREF _Toc260187493 \h 29NI 6008 USB PAGEREF _Toc260187494 \h 30Outback GTFX2524 Inverter PAGEREF _Toc260187495 \h 31Southwest Windpower Air X 400 PAGEREF _Toc260187496 \h 32Display Setup PAGEREF _Toc260187497 \h 33Executive SummaryWith renewable energy becoming an increasingly popular source for our changing energy and environmental needs, Iowa State University must do its’ part. To show the universities commitment in conserving and minimizing impact on global climate change, Iowa State University does it’s part in wind energy research.Our team plans to assist in the design and implementation of this wind turbine system, as well develop it as an educational tool for the university. Dr. Ajjarapu wishes to use the wind turbine purchased by the previous team and simulate an actual wind powered environment inside the lab. This will include designing sensors for the system, a user interface, as well as a means to protect the system while utilizing the maximum amount of wind.Unfortunately, our team lost a member this semester: Luke Lehman. He received a Co-op offer from OPPD and we wish him the best.Problem StatementThis is a continuation project. The ongoing project involved the design of a wind turbine energy conversion system that can be integrated to electrical power grid in the Coover hall power lab. The generator is rated around 400W. The wind turbine was to be installed on the side of Coover hall, but upon further investigation, the budget would not allow this to happen. All the protection and control aspects of the conversion system become part of the design. The extension of the project was to include supplying a standalone load in conjunction with grid supply. After the previous team left, it was discovered that many portions of the project were not left behind, and some were either unnecessary or did not work as intended. The user interface designed by the previous team was lost, the charge control that was designed to protect the batteries was insufficient, and the AC sensors did not function properly. It was left to our team to redo many of these mistakes and unfortunate miscommunications for our own project.Another major point of concern is the wind turbines internal controls. We wish to control the way the turbine operates ourselves, to maximize the use of this project. The internal control that is our biggest concern is the regulation mode of the turbine, which completely shuts the turbine off when the battery bank is full. We wish to work around this control and implement our own.Operating EnvironmentThe wind turbine and control system are no longer intended for outside use, so all prototypes designed will only be designed for indoor operation. Although our designs will not include outdoor use, we wish them to expandable and easy enough to modify to do so.All components purchased by the previous team will be placed inside and with close proximity to the wind turbine, so transmission infrastructure will be minimal.Intended Users and UsesIntended Users: The wind turbine system will primarily be used by Dr. Ajjarapu as well as his research assistants. It may also be used in future Senior Design projects so it must be expandable.Intended Uses:The system will be used to simulate wind powered generation. It will be used to study the implementation of small wind turbines and the characteristics the turbine shows during operation. This system also simulates a situation where the turbine requires a battery bank as a backup energy source and the issues that come with that requirement.Expected End Product and DeliverablesWind turbine power for standalone loadThe wind turbine (complete with control systems) will be able to operate a standalone load of our choosing (such as an induction motor or resistive load)Wind power simulationThe wind turbine will be driven by an outside source that is easily controlled. This will be a 3-phase induction motor coupled to the drive shaft of the turbine. A stable and useable test-bed for the turbine will be the result.Speed, Current, and Voltage MeasurementsVarious measurements will be made and displayed through a user interface.Utilization of Wind PowerSome of the wind turbines internal controls will be turned off and replaced with our own means. These new controls will still allow the system to operate safely but more efficiently.RequirementsFunctional requirementsFR01 The turbine will generate a 24V DC output.FR02 The turbine will generate a 400W peak output.FR03 Wind power will be simulated by driving the turbine by some other mechanical means.FR04 The wind turbine will not turn off when the battery bank is full, instead, the wind power that is available will be utilized the best it can.FR05 The turbine will power a AC stand-alone load by means of an inverterFR06 A user interface will control the wind turbine output and display of various system measurements (current, voltage, turbine speed, etc.)FR07 Various sensors will assist in displaying battery current, turbine current, turbine speed, system voltage, and power levels.Non-functional requirements NFR01 All wiring and electrical work complies with university and state electrical codes and regulations NFR02 Battery bank is in controlled temperature and stable environmentNFR03 The battery bank is properly protected from over and under-chargingMarket SurveyIn today’s world, the need for alternative, environmentally friendly energy generation has never been greater. The Department of Energy (DOE) has set plans and taken the initiative to have wind energy account for 20% of total energy generation or 300GW by capacity in the United States by 2030. This would cause a huge increase in demand for more efficient wind turbine designs to maximize energy generation while minimizing losses.Small scale wind farms are also becoming more popular. Although most systems today are “off-grid” types systems, demand is growing for “on-grid” systems that will allow the user to give energy back to the grid and use the grid as backup.Resources and SchedulePartsIt was important that we kept to our allotted $250 budget for this project, since the previous team needed additional funds for the cost of the turbine, inverter, and battery bank. Following is a breakdown of part costs:Item Cost Coupling $112 Current Transducer $21 Stop Switch $16 Display Materials $15 Kikusui Power Supply $0 3-phase AC Motor $0 Total $164 HoursHere is a breakdown of hours spent by percentage of each member.Hours BreakdownScheduleThe following is the work schedule followed for the second semester of our project.Work ScheduleDesignConceptual Sketch/Update Block DiagramWind TurbineBattery BankControl SystemTest-Bed ConnectionGridStandalone LoadCoover SystemInverterOrConceptual Sketch Figure 1This is the original concept sketch for the first semester. To illustrate the changes the project went through, it can be compared to an update block diagram below.Adjustable Power SupplyOutback InverterAC LoadUSB NI DAQGPIB-USBBattery BankRPM SensorDC SensorsDC SensorsLabVIEW InterfaceAC MotorCouplingUpdate Block DiagramWind TurbineThe previous team chose to purchase a wind turbine made for this type of application. The wind turbine is an Air X 400 W unit.The Air X 400 unit is designed primarily as a means to charge a battery bank which may be used to power electrical devices. This project wishes to emphasize it as a primary source of energy, with a backup battery source. This means there are features of the turbine that are useful, but several that can be seen as detrimental to our design project:Battery Bank RequirementThe turbine requires a battery bank to function. This is not entirely undesired but may be an issue. Instead of power being directly used from the turbine, it is supplemented by the battery bank which is charged by the turbine. Regulation ModeThe turbine can sense the voltage level of the battery bank. When this voltage level is too high (the batter bank is fully charged), the turbine enters a regulation mode. This mode cuts the output of the turbine until the voltage of the battery bank reaches a lower level. This is undesirable, given that we wish to make use of the wind whenever is available, not just when the battery bank needs charging. Thankfully, this mode can effectively be turned off my manually setting the voltage at which the turbine enters regulation mode to one that is so high that there is no concern that it will be reached. This is done by using a DC power supply connected to the leads of the turbine and turning a potentiometer on the casing.ControllabilityThe controls that allow the turbine to function, such as the voltage regulation system, reside inside the casing. This removes the ability to control certain aspects of the turbine that may be useful to the project. The casing cannot be opened and details or schematic of the controls cannot be obtained from the manufacturer.InverterThe previous group also chose an inverter for their project. It as an Outback GTFX2425. This inverter will be used to transform the DC power from the wind turbine to useable 120 VAC power to connect with the grid at Coover Hall or a standalone load within Coover Hall. It is important to note that this is a hybrid inverter that is capable of both these scenarios. It does not require an outside 60 Hz reference signal if it’s not connected to the grid. Many inverters are strictly on or off grid.ControlsThe previous team designed a charge controller for the wind turbine system. The controller trips a relay on the other side of the inverter when the battery voltage reaches too low of a level. Our team found this to be unnecessary because of the inverter’s internal control that automatically cuts off the load once the batteries reach 21 volts. So this controller has been removed without any consequence to the system’s requirements.InterfaceThe previous team had design a full interface for use with this project. Originally, our plans were to expand upon interface, but when the semester began it was discovered that the team had not left the interface with anyone, even our advisor Dr. Ajjarapu. So it was necessary that we started from scratch.The user interface is designed within LabVIEW. This gave us a gradual learning curve so that we could have it operational quickly. The NI-6008 DAQ is used to bring in readings from various sensors. The interface not only has to display measurements, but also control the output of an adjustable power supply. Luckily, a library of LabVIEW functions were provided for our power supply so design for this portion was minimal.The design of the input portion of the display was done by implementing tasks in LabVIEW’s block diagram editor. Each channel that was read from the DAQ was designated to get the appropriate reading. Some of these readings require mathematical manipulation to be correct. The values are then displayed using gauges on the front panel.Interface Front PanelDC SensorsThe DC current of turbine and battery are measured using 2 DC current transducers. Both are of the same type made by LEM (LA 55-P). The DC current transducer is rated at ±70A. It uses a galvanic isolation technology in which the high current being measured is not in contact with the electronic circuit outputting the low current. This model of DC current transducer also has low power consumption with its terminal connected to ±15V and can handle up to 8-gauge wire. When used to measure a 50 A current, The current transducer produces a 50 mA as an output signal, operating at 1:1000 ratio between input and output currents. LA 55-P Current TransducerRPM SensorThe RPM sensor became a requirement of this project as of the second semester. This means all design, testing, and implementation had to be done within a short time period. At first the RPM was intended to be designed as a standalone unit that would calculate and transmit an RPM value to a PC using a microcontroller and RS- 232 standards. Through group and advisor discussion this solution was discovered to be too complex for our needs and resources (mainly time). The general design was then switch to a simple pulse generator feeding into the built-in counter of the NI-6008 DAQ. This solution is less accurate and is more limited on maximum RPM measurements. Since our turbine will reach a max speed of approximately 2000 RPM, and will only require an estimated RPM value, these drawbacks are acceptable.The RPM sensor is design in separate hardware and software components:HardwareTwo infrared LEDs are used to generate the pulse signal required for the DAQ counter. This is a simple design that is popular in proximity sensing among other things. A comparator circuit is used as the receiver and a directly powered LED as the transmitter. The two LEDs are positioned next to each other. When a reflective surface is in front them, a voltage drop is induced upon the receiver LED. The comparator circuit then enters the ON state and outputs a voltage depending on the Vcc used. A load with a LED is at the output to provide a visual reference for testing.The circuit diagram is given below, with actual resistor values used. An LM 358 operational amplifier was used for its’ good output voltage swing of 0V DC to Vcc-1.5V DC. The 0V output is important for the counter inside the DAQ to work properly. A Vcc of 5V DC is used.RPM Sensor Circuit SchematicSoftwareThe counter of the DAQ counts the falling edges it receives from the input signal. The processing of the RPM value is all done within LabVIEW. The counter of the DAQ is continuously sampled in LabVIEW and is reset every 3 seconds. The number that is read from the counter is divided by the elapsed time to result in a frequency. This is then multiplied by 60, resulting in the RPM.Due to rapid variations in the calculated RPM value seen during testing, a simple averaging method is used on the output to achieve a smoother and more readable display. This is done by filling an array with 20 values and displaying the average of them. This same method was used in other portions of the display that suffered from noisy signals from the DAQ. Although it results in an output that is slower to react to rapid changes, it is satisfactory to our project needs. RPM Sensor Block DiagramThree Situations of ConcernAs said before, the wind turbine has a built in charge controller that effectively turns off the turbine once the battery bank reaches a certain voltage. This means that if there is available wind, it cannot be use to supply a load because the battery bank is full. To illustrate what exactly this means, we have outline three situations where we would rather have the turbine keep running rather than shut off. In all these cases, we consider the battery bank completely full so we do not wish to charge it.Wind Power=LoadIn this case, when available turbine power is equal to the load demand, we would most definitely want to utilize the wind power that is available. All the power produced by the turbine will supply the load, with none of it left to charge the battery. According to basic KCL principles, all of the power will flow into the load.Wind Power<LoadThe same applies to the situation where wind power is less than the load. According to KCL, all of the available wind power will flow to the load, with the remainder needed coming from the battery bank.Wind Power>LoadThis is the case where outside action is needed. If wind power is greater than the load, than the remaining current will flow into the battery bank, which is full as we stated earlier. If no action is taken, the battery bank will be damaged from overcharging. What can be done, is a simulated version of pitch control. In larger turbines, the pitch of the blades may be adjusted to change the output of the turbine to an extent. In our case, we are driving the turbine with an induction motor to simulate wind power, so we may also simulate pitch control by adjusting the speed at which we spin the turbine.All these situations make the case for removing the regulation mode from the turbine and using our own control to simulate pitch control. Regulation mode has effectively been removed by adjusting its turn-on voltage to one which is higher than we actually wish the batteries to reach, which is about 27V.ImplementationInterfaceThere were three different functions we wanted to accomplish with the interface. The first function was to monitor the voltages, currents, power, and the RPM of the turbine rotor. This was done by using the output signals from a DAQ. The DAQ we choose to use was the NI USB-6008. It allowed us to input the voltages from the sensors into LabVIEW.In LabVIEW, we were able to modify the readings from the sensors to view the actual values of voltage and current. The modifications were multiplying the voltage readings by 4.9, and the current readings by 10. Because the values we were reading were DC, to find the power we simply multiplied the calculated voltages and currents. The block diagram for this part of the interface is below. The RPM part of LabVIEW was discussed in the sensor part of this document.Voltage, Current, and Power ReadingsThe next function was to control the three phase power supply that was used to control the motor to drive the turbine. The PC communicates with the power supply through a GPIB-USB cable. A LabVIEW library was provided to us by Zhongjian Kang. We utilized two functions of this library. They were the ones that turned the output of the power supply on and off and adjusted the voltage and frequency. We used the concept of VF control to vary the speed of the motor. This is where voltage and frequency are kept at a constant ratio, which results in a constant torque output as the speed of the motor changes. Since we need to stay with the ratio of 220 Volts to 50 Hertz, the function was modified so that both the voltage and frequency were set by a single input. The block diagram for the controls is pictured below. Power Supply ControlThe third function was the charge control. The charge control was done by if loops. The first loop would compare the battery voltage to a given value, in our cases 27. If the battery voltage is lower than 27 volts, the voltage supply would operate at the voltage the user set. If the battery voltage was equal to or higher than 27 volts then it would check the current. The current was compared to zero. If the current was positive, this means the current was flowing out of the battery. Thus the voltage supply would operate at the voltage the user set. If the current was negative, LabVIEW would subtract a calculated amount of the voltage. This would lower the turbine output to a level at which the current would begin to flow out of the batteries. The charge control is pictured belowDC SensorsTo measure the output current of the transducer a correction was made on the connections of the previous design team’s work to get the right measurements of currents. A 100 ? resistor connected to the ground terminal of the current transducer instead of a connection to current transducer terminals. The voltage across the resistor is measured and then connected using a NI DAQ 6008 to the interface which displays current values outputted by both turbine and battery. The transducer was soldered to a PCB along with the resistor. A copy of data sheet with all the specification of LEM current transducer is added to the appendix section. DC Transducer ConnectionsA simple voltage divider is used to monitor the voltage of the system. It uses a 100 and 390 kilo ohm resistor. This ratio cannot directly be used in our calculations however, due to the smaller than ideal internal resistance of the DAQ.RPM SensorThe designed circuit was put together on a bread board for testing and then soldered to a PCB. When this was tested several bad connections were found and fixed, but too many problems were found near the completion of the project, and the more reliable bread board circuit was used in final implementation.RPM Sensor PCBPower SupplyA Kikusui PCR-6000W 3-phase power supply was used to power the 3-phase induction motor driving the turbine. The power supply is rated at 6000W and has an adjustable voltage and frequency. Facilities modifications were needed to power the supply, as it needed a 220 VAC source.Three-Phase Induction MotorIn order to simulate wind in the lab with the use of our wind turbine, we are using a three phase induction motor connected to the turbine with the help of press-fit precision metal bellows couplings. The motor is a basic three-phase induction SIEBER motor type LS7/T. It is delta connected with a 0.37 kW rated power and 1.7 A rated current at 50 Hz. The motor operates at a maximum rated speed of 2800 rpm and has a power factor of 0.8 cos φ. The motor also operates at a nominal output torque of 2N and a peak output torque of 10N. Motor coupled with turbinePress-fit Precision Metal Bellows Couplings: The press-fit couplings are used to connect the three phase induction motor with the wind turbine. The couplings are model BK5 that are backlash-free and torsionally stiff due to frictionally clamped connection and axial pretension of the tapered press-fit segment. They are easy to mount and dismount and can be separated, as well electrically and thermally insulated and have vibration damping properties. The couplings are able to withstand a temperature range of -30 to 120°C (3.6 F to 270F) and speeds ranging up to 10,000 rpm. In order to be easily mounted to the shaft of the induction motor and the turbine, the coupling hub has been taped with a D2 bored of 0.625” and a size 18 thread.LoadTo provide a load to the system, four 75W incandescent light bulbs were connected in parallel after the inverter. A base load of 150W is always connected, with two additional increments of 75W controlled by light switches available as well. This provides a means to demonstrate the operation of the system as well as give another visual reference of the power produced.TestingInterfaceTo test if the set up would work to view the readings from the sensors, we hooked the DAQ to voltage supplies. The voltage supplies were set to values we expected to obtain from the turbine. This proved that our set up would work. As soon as the sensors were complete, we hooked up the turbine and the batteries with the sensors to begin testing. We found that the values of the measured voltages were not correct. This was caused by the internal resistance of the DAQ not being a desirably large value like a regular voltmeter would be. This caused some current leakage into the DAQ which through our readings off. This was an easy fix. We simply measured the actual number we would expect with a multimeter and figured out the ratio needed for LabVIEW to display a proper value. We changed the values for the modifications to multiply by 7.926. We then used the coupling to hook the turbine to the motor to test the motor controls. We ran into a problem. We were unable to run the motor or get readings. We found that the motor controls needed to be in a separate loop. This caused another problem. The charge control is part of the motor controls but needs values of the battery current and voltage. Since the values were in the other loop, LabVIEW only allows that value in that loop till it is ended. This put an end to a software solution. The only solution that time would permit was to do this function manually.DC SensorsWith the turbine and batteries connected, and the turbine coupled to the 3-phase motor an analog current sensor and the two current transducers were used to measure the output of the turbine. The DC sensors were connected to the DAQ and the measurements were verified in LabVIEW. When the analog sensor and LabVIEW values were compared, we found them to be close enough for peace of mind.RPM SensorThe hardware and software design of the RPM sensor were first tested separately, and then as one system. The hardware was tested using an oscilloscope and multimeter to test the output of the receiver when reflective surfaces were passed in front of the LEDs. The software was tested by connecting the DAQ to a signal generator set at various frequencies and verifying the results within LabVIEW. This is where problems were discovered concerning the readability of the display. These problems occur due to the method used when calculating a frequency from a binary counter. There is a spike in calculated frequency when the elapsed time used for calculation is reset, as well as variations between iterations in the counter. These problems were addressed within the software resulting in a much smoother plete SystemAfter each portion was completed separately, everything was assembled and tested as a whole. Most of the bugs that came up dealt with the interface within LabVIEW. When the separate portions were joined together changes had to be made to ensure they did not interfere with each other. These dealt mostly with loop placement in the block diagram and separating the power supply control from the sensor measurements.We then proceeded to test the voltage and frequency inputs to the power supply and the resulting power output and RPM of the turbine.Power vs. SpeedOur team thought it would be useful to test the turbines full range of use. A Power vs Wind Speed curve was given to us by the manufacturer, and we wished to see how they matched up. We also needed to check what settings for the drive motor were needed for specific power output. Below are a comparison between the provided power curve and the one we made through testing.Measured CurveSupplied CurveIt is important to note that there might not be a linear relation between wind speed and the RPM of the turbine. This can only be tested by attaching the blades to turbine, installing it outside, and measuring RPM and wind speed simultaneously.ConclusionIn the beginning of this project, we had very high hopes. There were a lot of different aspects we hoped to expand upon from the previous group, and a lot of interesting ideas we wanted to implement. Unfortunately, due to the budget constraints halting the installation of the turbine outside, and the loss of much of the previous teams work, many of these ideas were not realized. However, we are happy that we have provided a more professional and useable test-bed for the turbine, as well as a RPM sensor that may be used for a plethora of different projects in the future. Our only shortcoming that we expected to deliver upon was the full utilization of available wind power. Throughout this semester, many different solutions for this problem were brought up, but each of them failed to accomplish what we wanted. The final idea of simulating pitch control through the software interface is one that we are confident will work, but bugs in the design of the control as well as time constraints on designing and testing it have made it impossible to complete. All of this being said, our group wishes to thank Zhongjian Kang, Lee Harker, and Dr. Ajjarapu for all their help and support throughout the project.Operation Manual492 Project Title: Wind Turbine System Design and IntegrationTeam: Brandon Janssen, Luke Lehman, Kenneth Thelen, Hassan Burawi, Elsammani AhmedAdvisor: Dr. AjjarapuAuthor: Stephen CopelandObjective: The high-level objective of this project is to implement a way of producing energy from a renewable wind source which can then be integrated into the Iowa State University power grid. This will be achieved by adding various sensors and a user interface so that the energy being produced can be monitored accurately. Functional Requirements:Turbine will output a DC voltage of 24VDC Turbine’s output is 400W Inverter supplies AC power to a load Sensors are connected to a data acquisition device (DAQ), and displayed on a LabVIEW interfaceWind power is simulated with a motor coupled with the turbineAn adjustable power supply is used to vary motor’s input to turbine Implementation:So far the implementation of the wind turbine system includes an adjustable power supply, ac motor, coupling to the turbine and motor, DC voltage and current sensors, an RPM sensor, a GPIB –USB connector, USB NI DAQ connector for interfacing with Labview software, a battery bank, outback inverter, and AC load in the form of four light bulbs. The ac motor is attached to the turbine in order to produce the needed energy to power the light bulbs after the signal is passed through the outback inverter. As the energy is being transferred from the motor to the load, DC voltage and current sensors along with the RPM sensor placed next to the coupling ensure that the signal is not going to harm any of the connected components. Labview is used to obtain the reading from the various sensors in order to act as a user interface which allows an operator to optimize power production from the turbine itself.Setup:Couple AC motor and turbine on an adjustable table so that proper leveling of the connection is possibleWrap the coupling with electrical tape leaving a small portion visible to reflect light to the RPM sensorPosition the RPM sensor next to the couplingConnect the turbine to a stop switch in parallelConnect a current sensor in series with the stop switchConnect a 2 x 12V battery bank in seriesConnect a current sensor in series with the positive terminal of the battery bankConnect a voltage sensor (voltage divider with 100K ohms and 390K ohms) in parallel with the battery bankConnect DAQ USB across 100K ohm resistorConnect an outback inverter in parallel with the voltage sensorConnect a switch between the voltage sensor and the positive terminal of the inverterConnect the load (light bulbs) to the inverterConnect RPM sensor to a 5V DC power supplyConnect 15V DC power supplies to current sensorsConnect power supply to 220V plug-in on wallPlug DAQ-USB and GPIB-USB into the computerTurn on power supplyOpen Labview interfaceAssign the proper address to the GPIB as 1Ensure that output of the power supply is offAdjust voltage and frequency of power supply using the LabVIEW interface until desired RPM or power output is reachedTesting Observed:During the time that I was able to spend with the design team I was unable to observe any successful testing due to an issue with the computers within the lab. However, the group assures me that they have successfully carried out tests during which the system was able to generate the power required to turn a load of four light bulbs on.Critique:Strengths:The system is able to transmit power from the generator to the load.The sensors allow for ease of control over the entire system.Weaknesses:Not able to implement system in real world conditions.While the system is not able to be placed on the roof of Coover, it has still met all of the requirements according to the group. Through the use of the various sensors and interfaces they have achieved a reliable user interface, the coupling of the motor and turbine has been successfully implemented, and through the adjustment of a power supply the system is capable of supplying power to an AC load. Plan for further research on the system include the use of the designed interface and to simulate varying wind speeds over time to better understand its effect on the a load. AppendixLEM LA 55-P Current TransducerNI 6008 USB Specifications: 8 analog inputs (12-bit, 10 kS/s) 2 analog outputs (12-bit, 150 S/s); 12 digital I/O; 32-bit counter Bus-powered for high mobility; built-in signal connectivity OEM version available Compatible with LabVIEW, LabWindows/CVI, and Measurement Studio for Visual Studio .NET NI-DAQmx driver software and NI LabVIEW SignalExpress LE interactive data-logging software Outback GTFX2524 InverterNominal DC Input24 VDCContinuous Power Rating2500 VAAC Voltage/Frequency120 VAC 60 HzContinuous AC RMS Output20.8 Amps ACIdle Power6-20 WattsTypical Efficiency92%Total Harmonic Distortion2-5%Output Voltage Regulation± 2%Maximum Output Voltage50 amps AC RMSAC Overload CapabilitySurge 6000 VA5 seconds4800 VA30 minutes3200 VAAC Input Current Max60 amps ACAC Input Voltage/Frequency80-150 VAC 58-62 HzDC Input Range21-34 VDCWeight56 lbsSouthwest Windpower Air X 400Rotor Diameter46 in.Weight13 lbStart-Up Wind Speed8 mphVoltage24 VDCRated Power400 watts at 28 mphTurbine controllerMicro-processor based smart internal regulatorBodyCast aluminumBlades3-Carbon fiber compositeOverspeed ProtectionElectronic torque controlKilowatt Hours/Month38 kWh/mo at 12 mphSurvival Wind Speed110 mphDisplay Load Setup ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download

To fulfill the demand for quickly locating and searching documents.

It is intelligent file search solution for home and business.

Literature Lottery

Related searches