Tesla turbine from old hard drives and minimal tools

Build a Tesla turbine from 2 old computer hard disk drives using basic hand tools and a pillar drill. No metal lathe or other expensive fabrication machinery is required and you only need some basic craft skills. It's crude, but this thing can scream!

Tesla Turbines promise up to 92% efficiency of converting air or fluid flow to rotational energy and its use can also be inverted for use as a pump with exceptionally high efficiency too. With compressed air becoming recognised as a feasible form of energy storage, we can see this device in everyday life soon as a source of locomotion. Factoring the simplicity, robustness and resillience to ingress of this design and you have something ideal for pumping heterogenous fluids like sewarage or fluids with suspended particulate. As a pump, this device has an important role to play in the developing world. More about it here: http://en.wikipedia.org/wiki/Tesla_turbine

Step 1: Tools that you will need

1. Torx set of screwdrivers to disassemble the hard drives and to build the rotor
2. Circle cutter - get this from you stationary shop for 1.99
3. Engineering Compass - optional, you can use the circle cutter to mark the workpieces
4. Sheet-metal drill bit and a 5 mm drill bit
5. Half-round file
6. Hole file
7. Craft glue or hot-melt glue
8. Epoxy glue to bond aluminium (that's aluminum to you lot in the US/Canada!)
9. Gaffa/Bodge/Duct/Electrical tape
10. Hacksaw with metal blade to cut aluminium (see proper pronounciation in item 8)
11. Pillar drill
12. Compressor to supply the air to drive the turbine. You can also use a drinking straw and blow really hard 'till your eyes pop out.
13. Some spare hard disk platters (chances are that you will have a couple of attempts to cut the right shape of slots in them)

Comments

[nerd7473]

I am building one

[tinker234]

wow could i use this to power a go cart

[Dyte]

Very exciting stuff man. Do you think this system works under water? Also, how long do you estimate will an air bottle of say one liter last when constantly ejecting at 80PSI? A rough estimate is fine because I have no idea :)

[gerrit_hoekstra]

I don't see any problem with an air-driven turbine running under water - I should try it, actually. There must be some clever applications for this? As for your second question - this depends on too many factors, espeically build accuracy. Bare in mind this thing was built in a woodworker's shop with nothing more than a pillar drill and a selection of files and a saw. You can either trial-and-error this to answer your question of go do a very interesting degree in mechanical engineering (!) and calculate it.

[Dyte]

Yes well, my first idea when I read this was to use this as a lightweight engine for a lightweight boat... running on compressed air. That's why I was wondering if this was feasable; if I'd need 20 liters of compressed air to make it run for half an hour it would be a bad idea :-D. Any idea? Well I suppose the trial and error method is still the best road to follow here! Thanks!

[Mr.Clodex]

So? The air would make the "Boat" float with displacement rules (Allowing you to glide across).

[itjstagame]

I wonder if you used the original DC motor as a motor and turned this in to a pump, how much force you could get in air pressure.

Basically, if you used it as a pump you may be able to use a stream of air as propulsion instead.

[gerrit_hoekstra]

All but the very oldest of hard drive motors are stepper motors and not DC motors. It takes a fair bit of circuitry to drive a 4-wire stepper motor, but the simplest solution is to just hack the original hard drive's PCB.

That's the wonderful thing with this invention by Tesla: it works very well both ways. So turning this into a pump will work.

[Predator106]

I'm kind of stuck...I have at my disposal (at least what I want to use..) a 150 GiB WD Raptor hard disk and it has 2 platters built right into it (they're a bit smaller though..kind of unfortunate, but the spindle does spin at 10k RPM. So I imagine that will help me achieve what I am cooking up...could probably switch it out, into another enclosure if I wanted to though. not sure how well it would manage though...as in, if there's a big reason why that size of platter is used (perhaps that particular spindle can't handle it. though I suppose that's not really a showstopper).


Anyways, how can I tell if it's ceramic or metal without drilling into it? Is there a way to drill a ceramic plate successfully, without cracking or shearing?

In theory at least, the boundary effect would likely be a lot better with ceramic ones, too. That's because the old metal plate drive platters actually had an oxide coding (yes..rust) which..if you look at an electron microscope, is actually extremely bumpy/spikey vs slightly hilly (it's a huge difference, at that detail level).

So what hard disc drives have you used? The issue is that well..the drive I want to use is only about 2 years old. I'm not sure how "ancient" yours are though...

You say that some have ceramics, but not many that you've run into (iirc, that's what you said..maybe that was someone else ;) but which manufacturers..years, sizes, etc are these?

I wish there were a good way of being able to tell it, without cracking it. But I don't think e.g. a magnet would be a "test" of that sort of thing.

Thanks.

[Maxwell Yun]

That's weird. My hard drive platters were aluminum, and they were from 1999-2000. Weird how my platters were solid aluminum.

[gerrit_hoekstra]

If the disk flexes in your hand, it is metal. If it snaps when you try to flex it, it is ...uhm... broken :-)

The surface effect is approximately the same over a perfectly smooth or a slightly textured surface - the drag force is applied well above the surface. We are not talking microscopic wind -vanes here, like in a pelton turbine! So ignore the point about the slightly rough iron oxide.

Besides, you will see that a metal hard disk platter is one of the most perfect surfaces you have ever seen - press two of them together and see how much effort is required to separate them.

[Maxwell Yun]

Nice project! I am using this for my science fair project, and it is very useful! I have access to a CNC mill, so I modified your instructable:
1. I machined the hard drive case completely flat.
2. I placed an aluminum block on top and machined it to fit on top of the flat surface.
3. I placed a polycarbonate square on top as a covering.
Thanks for this awesome instructable!

I am from America.

[epicnoobpwn]

Since I'm using an old (10 years old) hard drive, the walls of the turbine don't cover the top hard disks. I'm going to machine a wood enclosure out of wood and mount the motor on it. Basically a wood version of the hardrive casing.

[gerrit_hoekstra]

Should look very nice when done. Post the pictures here!

[wildwabbit]

Great project, a friend of mine is planing on making a slightly bigger steam driven tesla turbine, using parabolic mirrors to generate the steam.

I was intending to use this project as to have a play around with the porting, to see if there was an optimum or preferable set out, I am interested to know what happens when you off set the ports.

I spent half the day building this project and it took 30seconds with my air compressor for the hdd motor to internally disintegrate, I have not the slightest clue as to what the rpm's were but it was screaming, the momentum almost threw it out of my hand when the motor ceased. The shaft is jammed tight i could not get it to even slightly budge with a big pair of pliers.

Oh wel, l just have to find another motor with a long enough shaft, as the couple of spares that I have aren't long enough for the four disc and spacers.

Warning to others, don't get trigger happy with the compressed air, or your fun will be short lived.

Great project great results!!

[alvin9861]

I have started redesigning a auto turbocharger into a tesla turbine. The turbo is designed to handle up to 60k rpms and that makes it use perfect. I have built a larger turbine chamber to use 12 harddrive platters. I got an small sump pump to provide continous oil flow. I am using a 5 gal plastic bucket for the oil.

[The_Tom]

i wouldn't want to hook this up to the oil pump on a car. it pumps somewhat slow and when you start creating back pressure in your engine your going to cook the oil and destroy all the crank bearings. I was just thinking about adding one to a coolant hose. it doesn't destroy as much when there's back pressure and it pumps more liquid faster. if it slows down the coolant too much the car could over heat though.... but there is a temperature gauge on most cars so you should be able to tell when it has gone above normal. i am definitely trying this when i get some time. good luck.

[gerrit_hoekstra]

Go!!

[zimzam93]

so is there anyway i can use this to charge my phone or something, or will it just spin and make a pretty noise, sorry im noew to this kind of thing

[doomsdayltd]

really for this to generate electricity, the turbine needs to bee attached to an alternator to produce electricity.

[gerrit_hoekstra]

The "alternator" in this case here is the motor in the disk drive.

[doomsdayltd]

well er i'm just stuck with words right now, what i meant was connecting the turbine to an electric generator

[gerrit_hoekstra]

You will need to extend the length of the rotating shaft to make a flange coupling to a generator. Some disk drives allow you to gently tap out the rotating shaft, so simply replace this with a longer shaft with a flange or universal joint - a piece of rubber tubing will do, I would think.

[doomsdayltd]

thanks

[gerrit_hoekstra]

See the other discussions on putting a rectifier on the terminals of the hard drive motor to generate electicity.

[gerrit_hoekstra]

You won a medal for this? Well done! Your future as an engineer is secured!

I found a spreadsheet somewhere (and I don't know where it is now) that suggested that spacing should be about 0.8mm for something that has the viscosity of air. The more viscous the flowing material (water is more viscous than air) the bigger the spacing should be. It seems to not be too critical, but too tight and you loose energy from 'squeezing' your fluid through the gaps, and to loose and you loose energy from not sufficient fluid coming into 'contact' with the disk platters. My guess is that spacing is directly proportional to fluid viscosity (I do not know much about fluid mechanics).

The size of the inlet nozzle is of little importance - your aim is to get all the fluid to flow over the surfaces of all disk platters. I have a rectangular inlet that is a wide as the set if platters, which i made by simply sawing a cut into the one drive chassis, such that the airflow is directed along the outer tangent (i.e. 90degs to the radius line) of the platter. This way you will get maximum 'contact' between the energetic air and the platter surfaces.

[Slicerr2]

um i am doing a a science fair project this year with a friend and was wondering if you could give me the exact specifications to find out the spacing.

oh and by the way this years project is an extension of my project last year using your design it won me a bronze at nationals.

[Slicerr2]

oh and the sizing and angle calculations for the inlet nozzle would be great to

[Slicerr2]

why do you need to put the holes in the disks? the boundry layer effect should work without them shouldend it?

[gerrit_hoekstra]

Your theory is correct, but the practicality is that the air needs to go SOMEWHERE after it has delivered its energy (it looses velocity) through the boundary effect. So that is why there are holes in order to vent the air off. These vents are as close as possible to the center so that the air has the some time to remain in contact with the surface and impart its kinetic energy.

I guess that the math for how big the holes need to be is very complex, but as long as you have holes at least, this will work.

[hanmant]

how we can mesur effieciency of tesla turbine? how we can genrate power ? means i want experimental set up procedure.

[geoslim13]

look up "pleech" here on instructables and modify it for the tesla turbine

[erwic]

Just a few quick questions. What happens if you increase or decrease the size? Do you know which shape will give the most torque? speed? Thanks in advance!

[gerrit_hoekstra]

The size of holes I chose was pretty much based on the size of tools (sheet metal drill, files) that I had available. I have seen some interesting non-round shapes on other turbines - a parallelogram seems very popular and could be achieved with a bit more filing work.

[skizzilini]

A few more questions: What is a hole file? Have you tried just drilling some holes in the disks? Is a Drill Press necessary for this step, or can this be accomplished with different (or simpler) tools? How did you Enlarge the hole in the Back Plate in step 5? Sorry if these are obvious questions.

[gerrit_hoekstra]

A round file is another word for a hole file - used for shaping the insides of holes. Ideally, a 1cm diameter hole file should be used to shape those radial slots. A half-round file is good for enlarging the inside diameter of a platter for the back plate. A drill press is nice to have but I am sure if you are careful not to bend the platters, a normal drill will do. You should clamp your work piece down before drilling into sheet metal, because when that drill bit binds into the disk platter, finger tips can fly!

[skizzilini]

Where did you get this "blanking plate" from? I can find no equivalent in the HDDs I have.

[gerrit_hoekstra]

Not all hard drives have a blanking plate. It is much easier to make from from perspex, which is what I have now done. See-through perspex looks very cool because you can see more of the inner workings - not that it complex of course, but some people just don't believe me when I explain how it works.

[matroska]

 Last pic, step 11: reminds of supply depots in starcraft 1.

[tissit]

If you don't have the exact torx and don't care too much about bolt integrity, it's possible (albeit cruel) to hammer in some, perhaps improvised, flat/chisel screwdriver and use that.

Also, having a ratchet wrench allows you to put a fsckload of torque on a torx bit. Some of those screws are really tight.