Building a Disk Turbine
A model disk turbine built from castings. The flared port is intended for tests in compressor mode.
I just finished building an experimental disk turbine. I machined it on a 7" by 12" Gingery lathe, using a milling attachment that I designed and built this winter. The turbine was built from aluminum castings produced in a charcoal foundry. The model does demonstrate the action of Nikola Tesla's original disk turbine, which was built in 1906 and patented in 1910. Yesterday, during initial tests I was able to spin the model up to 10,200 RPM with compressed air at 75 PSI pressure.
There has been an increased interest in Tesla style turbines recently with at least two engine building societies actively engaged on the world wide web. There is also a small but growing body of engineering papers, doctoral expositions, and Tesla turbine books, including a new one containing plans for a model published by W.M.J. Cairns.-- The Tesla Disk Turbine.
What is a Disk Turbine?
A disk turbine is distinguished form a conventional turbine mainly by the orientation and form of the blades. As the name suggests the, the turbine rotor (or runner as Tesla named it) is a stack of disks. A conventional turbine uses a more or less fan, propeller, or waterwheel shaped rotor, (depending on whether it is an axial flow or radial flow style). By contrast, a Tesla disk rotor is usually composed of a set of disks, and star shaped washers fastened onto a shaft. The disk pack is arranged with the washers and disks alternating. The washers create a space between each pair of disks. Holes or ports are machined into the disk pack around the turbine shaft. These allow the exhaust of the gas or steam medium.
Disk blanks showing 3 ports near center.
A disk turbine is a radial flow gas turbine. Radial flow means that the hot gas which drives the engine is injected around the periphery of the blades, travels in a circular or spiral path. It is not an axial flow turbine. An axial flow turbine is similar to an airplane propeller -- the gas passes in a straight line along the axis of the shaft.
The mechanical action which drives a Tesla style disk turbine depends on the adhesion of the fast moving gas to the disks as it travels in a spiral inside the disk sandwich toward the center port.
Adhesion of the gas to the disks is affected by several factors. Current thinking is that the smoothness of the disks is a big factor. Smoother is considered to be better. According to this theory, a rough disk surface tends to create eddies and turbulence, which converts the motion of the gas into frictional heat rather transferring that motion to the disks. Since the function of an engine is to convert heat into motion, converting motion into heat in the rotor is counterproductive. Therefore the smoother the flow, the more efficient the rotor.
Smooth flow over a surface is technically called "laminar flow." Much research has gone into the conditions needed for creating laminar flow in the study of aeronautics. The advantages of laminar flow affects not only turbine blades, but aircraft surfaces, boat keels and sails and many other areas of interest to engineers.
Another factor affecting the efficiency of disk rotors is the gap width. The ideal gap width would accommodate the effective thickness of laminar flow on each disk in the sandwich, but no more. Any flow that is not close enough to a disk surface to help rotate it wastes energy.
Page visits since 4/13/03:
The claimed advantages of a Tesla disk turbine are that it is simple to build, and that it has a high power to weight ratio. There has also been speculation that it is fuel efficient compared to conventional axial turbines and has self-regulating capabilities.
A Promising Lack of Results
Despite these projected advantages, there are no practical commercially available gas disk turbine engines in production. Many new advocates of the disk turbine concept believe that the engine has been ignored for nearly a century because of engineering conclusions reached back at a time when modern high temperature metal alloys weren't available. Initial tests of a large 675 hp Tesla style steam turbine early last century by the Allis Chalmers Company yielded disappointing results. The 60" disks stretched after achieving a rotational speed of 3,600 RPM. As a result, Allis Chalmers dropped the development project, and little more industry funded engineering has appeared since.
Building a Turbine at Home
On the other hand individual interest in Tesla turbine experimentation remains keen. I wanted to find out for myself how well a set of disks and washers with simple ports would spin. After reading the Cairns book and looking at the plans, it seemed clear that I would probably need a milling machine to cut out the disks and ports for a machine of my own. I already owned a Dave Gingery-designed lathe that I had built the winter before using a home foundry setup. I decided to build a milling attachment with a rotary table. This took a month and a half to accomplish, and when March rolled around, I was able to test the new rig on a disk blank. It was a great feeling to mill the first port on a Tesla disk, using machinery and stock that I had built almost entirely from scrap.
Milling a disk port with the new attachment
How Big Should it be?
While putting together the milling attachment I had a long period of time to think about the design requirements for the turbine. I looked at a lot of sites on the Web and read everything I could find on the subject. I also read books on model aircraft turbine engine design, which has advanced spectacularly in the last 15 years.
The first question was how large to make the disks. I have seen several websites featuring 9 and 10 inch diameter turbines constructed by individual experimenters. However when it came time to test these turbines, which according to their builders were sized to accommodate 40 to 100 horsepower, compressed air was used. Most home shops have a small air compressor powered by a 1 to 3 (effective) horsepower electric motor. Obviously there is a mismatch between such a turbine's requirements and the output of a shop air compressor.
(c) Copyright 2003, Stephen Redmond, all rights reserved