Technical Report: Electrical Engineering Student Information Full Initials & Surname: R. Bester Student Number: 206015259 Mailing Address for Reports: Cape Town Campus Mr. L Khetla Department of Co-operative Education CPUT – Cape Town Campus P O Box 652 Cape Town 8000 Project Information: Report No: P1 Ti tl e of Proje ct: Th e Assembl y and Te st ing of a 3kW Ax ia l Fl ux Ai r-co re Wind GeneratorFi el d o f Wor k: De si gn, Man uf ac turi ng , Assembl y, El ectr ical Te st in g a nd Fa ul t Finding Company Information Company Name: Electrical Machines Laboratory Department of Electrical Engineering Stellenbosch University Postal Address: University of Stellenbosch Private Bag X1 Matieland 7602 Street Address: Room E166 Electrical Machines Research Dept of Electrical Engineering Engineering Building Complex University of Stellenbosch Banghoek Road Stellenbosch 7600 Telephone No: 021 808 3890 Fax No: 021 808 3951 Email Address: [email protected]Name of Mentor: Pr of M.J. Kamper (M. Ing S tel l PhD(In g) St el l SMIEEE MSAIEE Pr Eng Tel No of Mentor: 021 808 4323 1
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
This system is intended to replace the traditional wind pump system commonly found
in the agricultural sector. This system consists of a mechanical piston type of pump
driven with galvanized steel blades. This system is highly inefficient because of its
rugged construction.
This project replaces that system and consists of a turbine generator which generates a
varying voltage and frequency supply at variable wind speeds, induction motor which
drives the pump and a pump which pump the water.
Before designing the generator for the system we need to look at the behavior of the
other components or parts of the system as well. At first we look at the specific pump
performance curves (see fig. 1) to see where the working points for this specific pump
is. The working points on the graph are where the system curve intercepts the various
pump performance curves at certain speeds. The system curve is the total pressure
losses made up of the distance the fluid must travel which include friction and fittings
used in the pipeline. Now that we have the working points we can calculate themechanical- and input power of the pump by using the following formulas:
Mechanical power:
Ph = ρgQH
Where ρ = density of fluid
g = gravitational constant
Q = low capacity
H = head.
Input power:
Pin = Ph/л
Where л = efficiency.
The pump power must be matched to points where the turbine delivers optimum
power.
Fig. 2 is a graph that shows the characteristics of a 3kW, 3-blade turbine at certain
wind speeds. Optimum power points are indicated on the graph where the turbine is
most efficient. The input power of the pump is plotted on the graph and is known asthe load line. We want our system to operate as close to the optimum points as
possible at certain wind speeds. In other words we are trying to get the load line as
close as possible to the optimum points. The load line is shifted by playing with the
ratio of the poles between the motor and generator. The system will never operate at
the optimum power points because of the power losses in the motor and generator.
After calculating the input power of the pump the generator and induction motor can
be analyzed which deliver the mechanical power of the wind turbine shaft to the shaft
of the pump. The generator and motor must be sized according to the maximum
operating conditions of the turbine. The maximum wind speed on the graph in fig. 2 is11m/s. The maximum wind speed we are designing for is 10m/s which equals to a
0.8 mm diameter copper enameled grade 1 wire is used
After all 18 coils were wound; they were packed into a previously used mould in a
circle. 6 Coils was connected in series to with Farrell’s to form a phase. The 3 phaseswas then connected in a star configuration. This connection diagram is shown in fig 6
The next step was to reinforce the epoxy with weaving fiber-glass cloth. The mould
was closed up. The epoxy was mixed according to specific specifications and was
poured into the mould. After 24 hours the mould is set and is ready to be baked in a
oven. The dried stator is baked still in the mould at 100 degrees Celsius for 4 hours
and after that the heat is increased up to 110 degrees Celsius for 1 hour. The mould is
left to dry overnight. The next day the stator is taken out of the mould and inspected.
Part of the inspection process is to do a few measurements to check if everything is
correct. I used a multimeter and measured the resistance of each phase and I used a
LC meter to measure the inductance of each phase. Measurements were taken line toline and from line to neutral. The results are situated in table 1.
The phases are also connected to a megger to ensure that there are no short circuits or
leakage current between the phases. The completed stator is shown in fig 7
After the stator passed the inspection process, the generator assembly is started.
The permanent magnet rotor and is situated on both sides of the stator. A 1 mm air
gap is left between both sides of the stator and the rotor. The rotor and stator needs to
be lined up so that a 1mm air gap is present across both sides of the stator and is
uniform all around. This is done by inserting 1mm aluminum spacers between the
rotor and the stator and aligning them up. This part of the assembly is shown in fig 8.
Testing
After everything is aligned, the assembly is completed and more tests are done to
ensure that the generator gives out the correct voltage and that all 3 phases give a
sinusoidal wave which is 120 degrees apart from each other. This test is done by
connecting an oscilloscope to the phases and measuring the output voltage when the
generator is turned by hand. The results of the oscilloscope are shown in fig 9. After
the generator passed this test, I designed a control circuit which senses the output
voltage and frequency of the generator and uses these values along with contactor to
regulate the system and engage the load at the correct voltage range. If the voltageexceeds or drops below this voltage range, the load is disengaged. The system also
picks up the speed by sensing the frequency and breaks the system with resistive
braking if the generator were to spin out of control, bringing the speed down to the
regulated voltage.
Control Circuit
The load is engaged when the generator produces 170V.
The load is regulated between 140V and 280V. If the generator exceeds or falls below
this range, the load disengages. The frequency is monitored and kept between 20Hz
and 48Hz. If the frequency exceeds 51Hz or above, a contactor engages the resistive
braking which brings the overall speed down to a safe and manageable amount.
The control circuit and the power circuit are shown in fig 10 and fig 11.
I designed the cover using a program called Autodesk Inventor. The drawing was then
sent fabrinox in Paarl to be manufactured. This drawing is shown in fig 12 and 13.
Erecting the generator onto the tower
After the control circuit is in place, the generator ready to be put on its tower. The
tower consists of a head on which the generator is mounted, a tail which keeps the
generator pointing into the wind. Attached to the generator are 3 fiberglass 1.8m
blades. These blades weigh only 6 kg each and are light enough to pick up speed with
the smallest of breeze.
The tower is designed to fold in the middle so that the generator can be assembled on
the ground and pulled up using a winch and rope.
This is shown in fig 14, fig 15, fig 16, fig 17, fig 18 and fig 19.
After the assembly is completed, the generator is connected to the control circuit. The
2.2kw load (3ph induction motor) is also connected to the load. Test was done earlier
in order to test the performance of this system and is show in table 2.
This setup is shown in fig 20.
A weather station is also mounted on the pole which senses and records wind speed
every minute and saves the data. A flow sensor is attached to the system and will
show how much cubic meter of water is pumped up to date. These data is going to beused over the next 6 months and we are going to establish the efficiency of this
system by comparing the amount of water pumped at which wind speed and
conclusions will be made after that
Fault finding
After some testing we ran into problems when the wind generator stopped. I was sent
to investigate and to find the fault in the system. I used a DC constant current supply
and I connected it between two phases. With this I measured the voltage and I
measured the current. With these two values I determined the resistance of all three
coils with the formula R = V/I
I discovered that there was a short circuit in two of the phases and that is the reason
why the blades do not turn, because a short circuit brakes the system electrically.
The reason I used a Constant current DC supply is because if you use a multimeter or
a megger, the readings is unstable and difficult to read due to the fact that the
generator rotates slightly and generates a voltage. The DC supply brakes the generator
and makes it stable enough to make accurate measurements.
We took the generator down from the pole and tested the generator and found the
there are no faults on the generator. Then I measured the cable coming from the
generator head and found that there is a short circuit between the blue and yellow