A few years ago when hard drives were very expensive, I had to replace the hard drive in my Pee Cee.  I couldn't bear to throw away the dead (and expensive) drive so, per my usual habit, I found a nice place to store it in case I ever needed it again (I have a lot of stuff like that!!).

Anyhow one day, having nothing better to do, I dusted off the old hard drive and took it apart.  There's a lot of neat stuff in those gizmos like high quality small ball bearings, disks and such.

After looking at the pile of parts for a while, I decided to put these and some of my other 'stuff' to work doing nothing at all.  Here's what I came up with.  I call them electromotors.

For it's power source, this electromotor uses a push-type solenoid out of my junkbox.  To make it go, the connecting rod pin on the 'flywheel' has a magnet hot-glued to it.  The magnet passes the glass enclosed reed switch (a reed switch has two contacts that close when a magnet approaches).

When the magnet passes the reed switch, the switch closes for an instant and gives the solenoid a jolt of 12 Volt DC power. It will run for over a day from a small Gel-Cell battery or forever from a 12 Volt 'wall wart'.  The frame is made of soldered-up printed circuit board material.   At maximum speed, it turns at around 200RPM and makes just enough power to keep going.  The whole thing is about 8 inches long.


Here is a newer electromotor.  There is a magnet hidden within the stack of hard drive platters (the flywheel).  The magnet activates a Hall-Effect solid state magnetic sensor that is in the closed box in the middle.

The Hall-Effect sensor output is amplified and fed to the pull-type solenoid taken from an old IBM Selectric Word Processor (Made before the Personal Computer came out).  This electromotor also uses parts from a dead hard drive.

The rod 'big-end' and 'wrist pin' bearings are a couple of the precision ball bearings out of the drive.  I used copper wire for the connecting rod and, again, copper clad printed circuit board material for the framework.  Power comes from 120 Volts AC.  This one, the smaller of the two, is about 6 inches long and goes at about 250 RPM.


Movie Of Stirrer

This one actually does something!  It'll mix the oil back into a jar of natural peanut butter in about 24 hours.  Also made of junk, this one has a walking beam and a linkage to an escapement to run a couple of gears.  The output speed of the motor is about 180 RPM and is reduced to around one revolution in 4 minutes by the gears.



Movie Of Wibbly-Wobbly Machine

This one operates using what I'd call a modified Scotch Yoke mechanism.  As the flywheel turns, a ball bearing runs in the slot of the rocker which pivots on the bearing at the bottom.  On each end of the rocker is a linkage that goes to a solenoid.  There is a magnet mounted on the back of the flywheel and a couple of Hall Effect sensors, one for each solenoid.  It runs at about 500 RPM.

6 October 2006:  


Movie Of Tick-Tock

Having nothing else to do for a couple of days, I came up with this pendulum motor that uses the head position actuator from an old P.C. hard disk drive.  The original purpose of this motor was to make another peanut butter stirrer.  The ratchet assembly turns the fiber gear wheel but not with enough force to spin a peanut better jar so this one, at least for now, is just a motion-maker.



In order to make it run efficiently, I devised a simple electronic circuit to work with a sliding ball shuttle switch.  As the pendulum swings toward the left, the ball is "caught" by the support bearing on the left and the shaft slides through it until the end of the stroke.  When the pendulum swings back to the right, the ball shutles over to the right and grounds the contact rod.  The ball then stays against the contact rod as the shaft slides through it to the other end of the stroke.  As soon as the pendulum starts back toward the left, the ball moves away from the contact rod and un-grounds it, the positive going voltage triggers the circuit to drive the actuator coil.  The circuit times to turn on the actuator for something less than one half of the stroke of the pendulum.

This particular pendulum runs at about 80 full swings per minute, making the period about 750 milliseconds.  To power it for an entire half stroke, the coil needs to be energized for half the period or about 375 milliseconds.  With this particular "motor", I can drive the coil at substantially less than 375 milliseconds.  If I give it full drive time, it hammers at each end of the swing and runs fast because of the springiness of the pendulum arm.

In the unlikely event that someone would want to replicate this motor or something like it; I've photographed the schematic of the electronic driver circuit and photographed the circuit itself.  The potentiometer adjusts the "on" time of the circuit.  The time interval is adjustable from a few milliseconds (way too short) to about one half second (a little too long for this particular pendulum).


The switch on the upper left on the schematic is the sliding ball contact.  The integrated circuit is a CMOS (Complimentary Metal Oxide Semiconductor) MC14584BCP (one of several numbers including the number "4584") hex Schmitt Trigger inverter.  These are common, have a wide operating voltage range and work well in this application.  Also really nice is the fact that the input resistance is extremely high (a jillion Ohms or so), this allows long R-C time constants and very low input current requirements (ninnynanoamps).  To power this one, I dredged up a 9 Volt, 500 milliamp DC "wall wart" out of the junk box.

The 47K resistor limits the discharge current from the 0.1 microfarad "debounce" capacitor and the 1 Meg resistor allows the capacitor to charge relatively slowly, on the order of a few tens of milliseconds.  In the next stage, the 1.0 microfarad electrolytic, the diode (1N4148 or similar), and the 1 Meg potentiometer comprise the timing circuit.  CMOS IC outputs and inputs can be paralleled.  The 5 remaining Schmitt Trigger inverters are used to provide a sharply rising and falling drive signal.  Since each of the inverters can source more than ten milliamperes, the total drive for the coil can be between fifty and one hundred milliamps.  The 560 Ohm resistor in series with the coil is to limit the current the coil draws and the diode across the coil (1N4004 or equivalent) is to catch the glitch that is generated when the drive signal goes away.


3 February 2010:

The New And Improved Peanut Butter Machine

With Wibbly-Wobbly Power

Last year, a friend talked me into trading the original Peanut Butter Machine so, since then, I've been forced to toil away with a knife stirring the natural peanut butter that I like.

Enough is enough!  Since I'm fresh out of solenoids, the original Wibbly-Wobbly machine sacrificed itself for the cause.  I used the solenoids out of it to make a new Wibbly-Wobbly machine and then used a stack of 5-1/2" hard disk platters and some junk plastic gears and other stuff from the junque pile and, voila, the New And Improved Peanut Butter Machine.

The way it worked out, it turns the jar one turn in about an hour.  After a day or two, the peanut butter is nicely stirred-up and ready for me to enjoy.


Ball bearings from old Winchester disk drives were used extensively.

Escapement and gearing.  An "O" ring out of something or other is used as a belt.

Movie of New And Improved Wibbly-Wobbly Powered Peanut Butter Machine


Rush back HOME

Boy!  This is fun!

[email protected]

Visits Since 8 February 2010

vBulletin tracker