Hard Disk Generator (Version 1.oh)...

another "proof-of-concept" project...

Winter 2005

Note: This Project is fundamentally flawed in many regards (it may be interesting or even mildly amusing) but for plans and procedures for the "Working" Hard Disk Generator follow the following links... (addendum May 2008)

Human Powered Recumbent Generator , 2, 3, 4

This one of the smaller piles of Hard Disks that I receive a couple of times per year.  

My 2 main sources are a local computer recycler that has saved me upwards of 200 drives in a single load, and a pair of chums that work in the I.T. Sector and save me what ever goes bad on their network.

This load represents about 2 weeks of evenings stripping down the drives and sorting the scrap into various piles.

The Aluminum frames are recovered for the foundry and are an excellent stock as they flow nicely and are easy to machine.

The yellow bucket of "Platter Hub Motors" needs to be melted twice, once to reduce the Alum from the internal metallic parts, and the second time for what ever project is underway.

The blue bucket is overflowing with "Rare-Earth" magnets that have also been recovered. These are still stuck to a steel backing plate, that I'll remove on a slow day when I need some truly mind numbing work.

Above to the left are what the magnets look like with their backing plate still intact. To the right are the "Cleaned" magnets ready for this project.

The magnets initially were sorted by shape and polarity. The faces of the magnets are polarized, but there are pictures below that better illustrate this.

The strength of these magnets is amazing, especially when they get stacked and you want to get them apart.

There are numerous ways to remove the magnets from the steel back plate, but I found that clamping a corner of the plate in a vise and bending the plate with a pair of vice grips was the quickest and resulted in the least number of broken magnets. While sharply rapping the edge of the magnet with a hammer was the worst method that I'd tried.

With a reasonable stock of magnets on hand, I laid-out a template for 6 of the more common shaped units, and milled it into a 1/4" birch plywood board.

This will be the base of my test-rotor for the HD-Generator.

The pockets are recessed into the rotor such that the magnets can be easily glued into place and not be able to attract to each other and make a jumble pile.

The center is drilled out to 3/8ths" and centered relative to the magnet pockets.

These are the magnets that I opted for, and were the original shape that I scanned when I laid out the CNC pattern.

The magnet that is laying flat on the mill table was placed across the T-slot so a screw driver could be applied to pry it loose, otherwise it would have to be slid along the table to the edge before it could be lifted.

Each magnet is approx 1.5" long, 0.5" wide and 1/8th" thick.

The magnets are glued into place and left on a machined flat cast iron table top to ensure that everything remain flat.

The birch was reasonably flat to begin with, but now is doubly so.

This is a cache of dead motor stators that I have been saving for sometime in anticipation of this project. 

The wire is solid copper core with a light enamel coating. The donors pictured here range from  dead drills, circular saws and even a 3 speed fan motor.

Although using recycled enameled wire is frowned upon by some as the insulating coating may have a knick and cause a short, for this preliminary series of tests I feel confident that it will be fine.

I carefully read a number of sites and all seemed to agree that the center of the coil should fully straddle the magnet pole that it is going to pass over.

This was the premise I used to draft the parts for this coil winder. The idea is that the winder will be able to make various sized coils by stacking and changing the core.

With that in mind I opted to spend a bit of time making the coil-winder rather than just using a length of pipe or a screw driver handle.

To the right are 3 coils that were wound using AWG #24 (0.021" dia).

The coils are 75, 100 and 125 turns respectively. These values and the gauge of the wire were selected somewhat arbitrarily, though the space available for shoe-horning coils over the magnets was also a consideration.

Shifting gears yet again... The image to the left is of a small pile of magnetite or "Black Sand" that I sifted from my pile of foundry sand.

This material will be mixed with a resin and catalyst to boost the efficiency of the magnetic flux to electricity transfer in the coils.

Above to the left is the back of a cookie pan that I liberated from our family kitchen that has been liberally coated with magnets. To the right is the inner cavity that the sand is applied to.

The sand is poured and shaken in the pan, leaving the magnetite adhering to the surface. This is the 3rd incarnation of this design and finally yields a reasonable amount of material for the effort expended.

The black sand is simply rubbed down the pan past the edge of the last magnets and is poured into a container.

As a word of caution, on my first attempt I sifted the sand for about 15 minutes without any respiratory protection and was shocked at the heavy viscous mass of fine sand that I later blew from my nose that evening... For my part I resolve to use  respiratory protection. ( was going to include a picture but thought better of the idea upon some reflection.)

This image is of the magnet field pairs of the individual magnets as they are arranged on the disk.

Each magnet has 2 cleanly defined poles on it's face. This image was taken using a sheet of paper placed over the rotor and sprinkling a small quantity of the magnetite on the paper.

The image above to the left is the rotor stuck onto a saw blade, with a rim of tape around the outer edge, and a 3/8ths drill bit (also taped) stuck in the center. Once the resin and black sand mixture is applied, the sheet of paper will be placed over the rotor and the second saw blade will be pressed down until the magnets get a good hold on it, ensuring a thin flat rotor.

The image above and to the right is just that, as well as 2 of the 3 coils filled with the same mixture.

The magnetite was mixed 5:1 by weight to the resin. This was the maximum concentration that I could achieve and still have resin semi-flowable.

The concoction was left to cure for about 2 hours and appears to your immediate left.

The 3rd coil was not filled as that was the limit of the materials that I had mixed. 

The rough rotor is pictured to the left, chucked onto a short length of 3/8th" threaded rod in the drill press. After a couple of minutes with a rasp the trued and rounded rotor is on the right.

The magnets are now completely embedded in the resin/black sand mixture, but again the outline of their field is visible due to the dust that has found it's way onto the face of the rotor.

The drill press is belted down to 220 RPM which is the speed that the current wind generator (Version 2 PM motor unit documented elsewhere on this site) has been hitting fairly consistently whenever there is a breeze blowing.

The images above show a second layer of magnets over the originals, as well as a third layer to the left.

The data below are the results between the various coils and the addition of more magnets.

The data above is measured on the AC scale... The value recorded is the peak value that I was able to hold for at least 3 seconds and repeat at least once. The reason that I'm making the above disclaimer is that there was a huge fluctuation between the values read and the relative position of the coil over the ring of spinning magnets.

A 4th coil un-cored of 30 turns using a 0.031" dia wire was not included as it consistently registered less than half of the 70 turn un-cored. The physical size of the thicker wire would have made the coil a monster (larger than the coil winder could hold) to hit the 100 turn mark.

So for now, it looks like I'll build a larger rotor to hold either 12 or 16 magnets in a ring, with a second layer atop those. This will be embedded in the same resin/black sand mixture. 

As to coils, I may have to buy a few spools of #24 enameled wire as I doubt that I have enough wire of a consistent gauge on hand to wind a dozen or so 200 turn coils. These will have a marginally larger core and be at least another 1/4" thicker. The idea is that each coil be able to produce over 1.0VAC (ideally 1.25VAC).

The additional magnets around the larger diameter rotor should contribute significantly, as well as the additional 75 turns on each coil. With 12 coils @ 1.25Vac and wired in series the effective output should be close to 15VAC, once rectified to DC be just over 13VAC.

Although this feels like one of the longest pages that I've posted in quite some time, it does cover a significant amount of information that is all very basic to the progress of this project.

In summary the original "Stepper - Wind Generator" actually produced power, albeit limited to 1 or 2 watts on a really windy day. 

The "PM - Gen (Ver. 2.0)" Was easily a tenfold increase but limited to charging 6V and 8V golf cart batteries as it rarely gets windy enough on a consistent basis for reliable 12V charging.

The aim of this project "HD Gen" is to make at least a 100Watt unit that can be easily duplicated.

My many thanks to the numerous web-sites (Specifically Hugh Piggott and Otherpower.com as well as the numerous poster's that have shared a wealth of valuable information)... there really isn't anything original in this dispatch beyond the experience that I gained getting all the various parts together. 

Hard Disk Generator , 2, 3, 4, 5

Back to Energy

Disclaimer (an unfortunate necessity)
All Rights Reserved theworkshop.ca © May 25, 2008