On Aug 17, 2009, at 3:42 PM, Kyle Mcallister wrote:
V, and Horace of course,
Got out the original ball bearing motor which works, and connected
a high current transformer of indeterminate turns ratio (made it
myself years ago, do not remember). Technical details of the
transformer are below:
Core: several spools worth of iron rebar tie wire wound in a donut
shape. Maybe a foot across.
Primary: 10AWG THHN stranded. Number of turns...a lot.
Secondary: 6 turns of "some heavy wire". Really, half of a pair of
auto jumper cables.
Visual appearance: looks like a chocolate donut with some red jelly
leaking from one side. Suggestion... use to attract cops if broken
down on side of road. But one half of a jumper cable set won't do
much, will it?
If you measure the open circuit output voltage Vo, and since you know
Vi = 120 V, you can calculate the number of primary turns Tp from the
proportion:
Vo/Ts = Vi/Tp
Tp = (Vi/Vo) * Ts = (120 V)/(Vo) * 6
Tp = (620 V)/(Vo)
Example: if Vo = 200 mV, then the number of primary turns is:
Tp = (620 V)/(0.2 V) = 3100
Primary goes to 120VAC. Secondary across the BB motor.
When run normally (with flywheel) the motor will spin in either
direction, and will end up steady at maybe 150 rpm. Heating is not
too bad now. I do need to tweak the turns of the secondary to find
optimum "feed" for the motor. I do have the temptation to run
240VAC across the primary, as my lab is wired for a rather robust
bit of that.
From the low rpms it sounds like the output voltage is too low for
the resistance of the motor. Your transformer should have plenty of
power at 120 V. If you have room in the core you could add some more
turns of secondary wire to get the secondary voltage up. Also, it may
be that weld dimples are preventing the BB motor from conducting
except sporadically. Checking performance of some new bearings might
be worthwhile. I found that, for oil lubricated bearings, soaking
overnight and washing in gasoline and then blowing just a very little
bit of powdered graphite (sold tubes in the locks section of box
stores) and working the bearings works well to reduce arcing. The
graphite comes in visible sized chunks (too big). They have to be
crushed to a fine powder within the bearing cages. I do this using a
Dremel tool or drill with a small buffing pad to spin the bearing and
wear in the graphite.
Interesting and unexpected thing: when the flywheel was removed,
and current was applied, I tried to see if the motor would spin
with just the shaft. It doesn't...but there is a torque produced.
If you do it right, the shaft will oscillate back and forth,
CW...CCW...CW...CCW...etc. maybe 60 degrees. Back and forth, back
and forth. Thermal oscillation of some kind?
Sounds like you might have some weld dimples on the races. When you
move the shaft manually do you feel cogging? You might need new
bearings.
Or not getting over a 'hill' of mechanical resistance at some point
along the bearing's race?
Yes.
Either way, it wasn't expected, and was fun to play with.
Horace, if you think it warrants it, I'll tweak the transformer to
get it working more optimally. I do have some high current
rectifiers. Can probably rectify the juice, if needed.
--Kyle
My feeling at the moment is that the most important thing is being
able to take measurements. No understanding can be achieved without
measurements. We already know from videos and personal experience
they work on AC and DC and bidirectionally. Its *how* they work that
needs nailing down.
The most important measurements are voltage drop across the motor
when running vs stopped. It is also important to be able to measure
current through the circuit at the time of the voltage measurements.
These measurements are probably best made using a battery for power,
and while having a high rpm BB motor. The run time doesn't have to be
very long to get these measurements. So, even though using a battery
is a pain, because you have to recharge it, and battery life is
likely very short, it makes things much easier and simpler to get the
data needed to have any understanding at all of the principles of
operation of the motor.
Measuring the voltage across the motor is fairly simple. You just
measure it. The difficult part is setting up a current measurement
resistor that has enough ohms to get a good voltage reading across
it, and which doesn't heat up much, changing its resistance, and
which doesn't have so much resistance it reduces the motor current
too much. The alternative is finding an amp meter that can measure
the current directly. The current sense resistor is way better,
because you can put a scope across it and see what is happening to
the current in real time.
The other hard part is measuring the current sense resistor's ohms,
Rs. One way to do this is to find a larger resistance, R2, that is
easy to measure accurately, say a 25 ohm spool of wire. You just put
the two in series and measure the resistance Rt through them. Then
you put a battery across the series and measure the voltage Vt across
the two, say 12 V. Then,while the voltage is still Vt across the
pair, you measure the voltage Vs across the current sense resistor Rs.
The current It through the pair is:
It = Vt/Rt
The resistance of the current sense resistor is then:
Rs = Vs/It
Example: the resistance of the pair in series is 25.1 ohms when the
voltage across them is 12 V. The current It through them is:
It = Vt/Rt = (12 V) / (25.1 ohms) = 2.09 amps
The voltage across the current sense resistor is 10 mV.
Rs = Vs/It = (0.010 V) / (2.09 amps) = 4.78x10^-3 ohms = 0.00478
ohms
More importantly, you now know that you get It/Vs = (2.09 amps)/(0.01
V) = 209 amps per volt through the current sense resistor. If you set
your scope y axis for .5 volts per division and put your ground and
probe on opposite sides of the current sense resistor you will see
about 104 amps per division on the scope for that probe.
If you use A/C to drive the motor then you will need a 2 channel
scope to make any sense of things. Also, the situation using AC is
far more complex, but the good side is you might just find out some
things about the motor's dynamics you would not know otherwise.
Best regards,
Horace Heffner
http://www.mtaonline.net/~hheffner/
