Bent Genny
Part # 4 - Single, Blank & Dual Rotor
Comparison...
Jan 17th 2007
In concluding
Part #3, a data collector was added to provide
accurate data relative to the RPM, Voltage, Current, Wattage and the total
wattage per logging session recorded. The data is sent via RS-232 to a PC that
stores the data for display and later analysis.
The
Genny is used daily by 3 of us within the household. My wife & I each like
to crank-out 30Watt/hrs per session. My goal is to slowly build-up to 50Watt/hrs
over the next month or so. I can appreciate that 50Watt/hrs may seem like a
dimensionless value if you've never tracked how much energy you expend over time
in an exercise regime (I couldn't), but let me assure you over the last few
weeks I have grown a considerable respect for the value of a Kilo Watt hour.
At 6
or 7 cents per KW/H here in rural Ontario, it has to be one of the best deals
going, even if it tripled, or was ten times or say $1/Kwh it's fantastic value
once you start to see how much work it really is.
With
so many data samples being logged I've revised the "Delta Stator"
graph with just a single Rotor in place.
The
graph above is comprised of 1229 samples of RPM & Wattage plotted in a
"Scatter Graph". The variability of the wattage output at any
particular RPM value is due to a couple of factors;
a)
the ADC (analog to digital converter) that is counting the stator pulses to
determine RPM has a limited resolution of 7.5RPM for a 16pole rotor.
b)
the instantaneous reading is not within the context of the rotor accelerating or
decelerating at the moment it is being sampled.
The
key data points are that current starts to flow into a 12V load @ approx 280RPM
and that the generator peaks out at 160watts @ 775RPM.
This
becomes the baseline or reference for any changes to the Generator.
Just
to be clear, my intention was never to build a Hyper complicated exercise
machine, but to build a basic platform that I could test Wind Mill generator
designs on.
To
that end a second 1/4" steel rotor is cut and fixed to the generator.
The
idea is that the permeability of the heavy steel plate draws the magnetic flux
lines more thoroughly through the stator.
The
second rotor is fixed in position relative to the first by 4 lengths of threaded
rod.
The
magnetic draw on the second plate is considerable, certainly enough to hold it
in place quite firmly (but I bolted it down just the same).
As
this was just an intermediate step I opted to settle for only 603 samples, which
make up the graph below plotting a similar trend as shown above.
So
what's the difference between a single rotor and a plain old piece of steel on
the other side of the stator?
Charging
starts @ 190RPM and the generator peaks out 183Watts @ 670RPM.
The
next step is to make a complete second rotor with magnets aligned to the first.
The
yellow "Witness Marks" are painted on to create a point of reference
for the magnet placement template (as used on the first rotor in Section
#1)
The
Magnets are arranged so as to attract the two rotors together, ie; N to S.
I
checked this 3 times to be sure I had good alignment before casting the rotor in
resin.
I
lined-up the first 48 matching magnets that I could find (3 per pole).
Although
the shape is identical, these are marginally thicker.
The
original rotor was set-up with poles that were 1/2" tall, where these
ended-up being only 0.4" tall.
I
settled for what I had immediately on hand rather than hunting about for 16
more.
Again
the rotor is cast with a polyester resin that has had blue enamel paint added as
a tint.
Although
I'm just learning the most basics of generator building, I can see eventually
standardizing the colour schemes of the rotors of various machines based on the
number of poles or variations in diameter.
The
mold that is shown is the same one as was use for the original rotor, and as
such can be considered re-usable.
Holy
Fuck!!! but do these rotors ever have a lot of drawing power between them when
being installed.
I
thought that I'd broken the stator the first time they slammed together.
There
is a slight edge below the surface of the poles to grab onto, I can easily see a
finger getting busted if one wasn't paying attention at this stage.
Bear
in mind that this is a comparatively small unit with rather sub-standard
components. If you were to use premium grade magnets from a source such as
"Otherpower.com" the need for vigilance would be increased (as would
the output).
Since
this is still largely in a test mode, I opted to carefully set the nuts shown on
the rotor bolts at a uniform height relative to the original rotor using the
calipers shown. Eventually I will cut a solid spacer that would be faced-off on
the lathe for truer alignment.
The
stator took a bit of fiddling to position between the two rotors with a
3/4" gap between the 2 rotor pole faces.
The
stator is approx 1/2" thick, and has 1/8th" gap on either side
(approx).
The
key concern I have is that the rotors aren't scuffing against the stator and
eventually causing a short or open within a coil.
The
graph below is made up of 1115 data samples, with a marginally higher cut-in or
charging speed @ 230RPM (possibly from the stator being aligned differently) but
with a peak output of 203Watts @ 550RPM.
The
generator as posted above may well still be improved with some tweaking here or
there, but I'm satisfied with it's performance for now. The "Peak
Output" is somewhat misleading in that it is actually the peak output of
the human that is running the pedals. This is certainly a limiting feature of
this configuration as "Test Bed".
In
conclusion, I feel more comfortable about the relative output capabilities of
the 3 Rotor configurations outlined above; Single, single with a blank Rotor and
the highest output from a dual rotor arrangement.
Also
I'd like to note that there may actually be a further boost in output by
dropping the "Black Magnetite Sand" that was embedded within the
stator. The idea of incorporating the magnetite into the stator was to improve
the drawing of the magnetic flux through the coils, while with a Dual Rotor or
even Blank rotor design it becomes redundant and even may degrade the output.