Concept Design Review Documents
Table of Contents
CONCEPT DESIGN REVIEW DOCUMENTS
This page contains the working documents used in creating the concept design.
In order to analyze the amount of wind power we can expect at RIT, average windspeeds need to be obtained. AWS Truewind windNavigator was used to obtain average windspeeds at various heights for RIT's campus.
Click here to view the PDF page of the windspeed results from AWS.
The WIND ANALYSIS file is an excel file that shows the breakdown of the total hours per year each windspeed value is expected to occur. The power in the wind is also calculated in this file by the equation Pwind = 1/2*density*area*windspeed^3. Power in the wind is a measure of the amount of power that we could expect to harvest based on the density, windspeed, and area of the blade diameter with a turbine efficiency of 100%. No wind energy collection systems can capture 100% of wind energy. Thus a turbine efficiency assumption must be made. In this analysis, an efficiency value of 20% was estimated (this is a low-end estimation). The power capture can then be found for each windspeed value in the table. The Power at Xm graph shows the amount of power we can expect to achieve from our wind turbine compared with the power in the wind. The variables highlighted Yellow can be defined by the user. So choosing a height of 3 m and a blade diameter of 1.5m, for example, we see that we can expect a power range of (2 Watts to 110 Watts) between windspeeds of 2m/s and 8m/s.
Power will be obtained by magnetic induction from a small generator motor. This generator will have a rotating shaft that oscillates magnets through internal coils. There are many types of generator motors with a variety of specifications concerning voltage and current output related to shaft rotation rpm. In order to charge a Lithium Ion battery, it is desirable to have a permanent magnet alternator (PMA) that yields a voltage output between 0-15 volts at an rpm value that can be reached by a wind turbine design or by a hand crank application. Through research, a PMA was found that is designed to produce a high voltage at low rpm values (between 150 and 1000 rpm). The link below will take you to the website where specifications of this PMA can be viewed.
The two most applicable PMAs from this manufacturer are the DC-540 and the DC-520 models. The DC-540 reaches 12 volts at 130 rpm. The DC-520 model reaches 12 volts at 275 rpm. An analysis was performed to view the comparison of the two PMA's. In this analysis, the voltage and current specifications provided on the website were used to model the power output that could be expected at a range of rpm values. The provided voltage and current charts were used to fit a regression equation to the data. The regression equations are displayed on the charts. These equations were then used to extrapolate voltage and current values at the rpm range of values in the table. Power output was found by the relation P = Vi. From the PMA Power curve chart, the power level output that can be reached by the DC-540 model is greater than that of the DC-520 model.
View the PMA ANALYSIS here.
Liion Battery SelectionWe have done the analysis (on Select Product Concepts) to use a li-ion battery and we decide a pack or we use a pack that isn't military grade or from a battery. A lot of the decision of the battery pack depends on how much power the groups are going to need. Find there is some risk having to design the protection for the li-ion Batteries. It is safer to have a designed pack(from a company) that isn't military grade or battery pack.
An Excel Spreadsheet is made up to help choice the Li-ion Battery. View the Choices for Li-ion Batteries here.
In the interface meeting. Long range decided that they would need this much power:
|Double Current for Margin||600mA|
|Lateral Surface Area||3751.6||mm2|
|1/2 Lat SA||0.001875782||m2|
A Flyback typology is what we are planning on using for our DC-DC converter. We will can have an input from 0 to 20V and we want to convert to 5.5 to 8.4V so we need a boost and buck converter. We also want to have an isolation barrier between the output and input. This is done through a flyback controller which uses a transformer to isolate the primary and secondary.
The controller that we are thinking about using is UCCx803. The x will be a 1 or 2 depending on if want the temperature range to be from -40 to 85C or -55 to 125C.
This controller is a PWM that creates a constant current. This controller is going to be controller by the micro controller. We are still deciding what the other parts are going to be.
Shield for Power Electronics
The power electronics can create a lot of noise because it is a switching power supply. We are looking into using a metal or conductive plastic casing. We found some plastic casing that we think would work well. We haven't decided on an exact box because we are not sure of the dimensions of the power electronics. We would like to at this point use Hammond Manufacturing. This is one website that we are considering using: http://www.hammondmfg.com/dwg3RFI.htm
ConceptOne concern brought up in the concept design section was that we have to much to do. To help eliminate that program we found that on national's website http://www.national.com/analog called Webench Designer. The Webench design program allows you to put in your input voltage ranges, and your output voltage and load current.
This will start the program that has every controllers that has that input and output ranges. It created a design for each controller. This interface allows you to look at switching frequency, lowest BOM cost, smallest footprint, and Vout peak-to-peak.
Once you picked your design we can know start analyzing the design.
Overall this gives a great product. We would consider handing this off to another group but our group has experience with debugging and modifying power electronics. This should cut back on our work and allows us to explore the windmill project more effectively.
We are going to be using WEBENCH from national to design our project. We are looking at controllers LM3100, LM22672-ADJ, and LM22680-ADJ. These have high efficiency with smaller board space. The main advantage of LM22672-ADJ and LM22680-ADJ is that both have full webench tools. The advantage of LM22680-ADJ has higher peak output current(2A) where LM22672-ADJ has only 1A peak current.
At this point we are going to go through a quick guess on which chip we are looking at for the interface meeting so that we have an idea of the size.
I am looking at the LM22680-ADJ chip and adjusted the parameters for what I think at this point might be good. The largest and highest part is the inductor. The webbench says that it needs to be between 23.31 to 46.1u with a target of 31.08u. The original part is http://www.bourns.com/data/global/pdfs/SDR0604.pdf. My problem with this is it is an unshielded inductor so I looked on the site for other inductors. I like this inductor better http://www.we-online.com/eisos/pdf/744066330.pdf. It is shielded, has higher inductance and lower DSR! Either way the different parts that are shown on the WEBENCH has nothing higher then 8mm so height shouldn't be an issue