More on Inlet Nozzles & Special Test: Tesla Disk Design Vs. Phoenix Hybrid Winglet Design

March 27, 2002

Last month we examined how various hydrocarbons perform under cold conditions. By vaporizing liquid fuels and passing them over an ignition spark we were able to determined which would be suitable for all weather conditions. We also took a brief glimpse at a couple of our fuel processor components.

We're going to start out this month's study by clarifying what makes the Tesla turbine operate and how components work together to produce an efficient design.

We'll also take a look at how geometry changes to the disk pack affect efficiency.

As stated by Nikola Tesla, his disk turbine is best used as the primary stage of a multi-stage system, followed by a Parson's type bladed stage. The Tesla design is more efficient in converting high pressure, lower volume kinetic gases into rotary power, whereas bladed turbines are more efficient at converting lower pressure, high volume gas into power.

A good starting point is around 125 psi - 150 psi of pressurized gas, air, etc. to feed to a disk type turbine. Between the supply of pressurized gas and the disk pack we need to insert a nozzle to convert static pressurized gas to a high velocity/high kinetic energy fluid. The best and most efficient way of doing this is to use a convergent/divergent nozzle.

Figure 1 shows a model of convergent/divergent nozzle similar to a design published by NASA Tech Briefs (January 2001, pg. 60). Pressurized fluid enters the nozzle from the left at subsonic speed. As the fluid passes the largest diameter of the insert, it accelerates to the speed of sound. Continuing its flow to the right, the fluid expands rapidly, exchanging heat energy for supersonic velocity. The NASA design is easier to build than a DeLaval nozzle since the insert is machined rather than the casing. An added benefit is that the outflowing fluid converges upon itself to the center of the nozzle rather than following the outer casing. 

For more information visit www.nasatech.com, Mechanics section, paper #KSC-11883.

After we achieve a supersonic flow of gas through the nozzle, the next challenge is to convert the highly energetic gas into work. Early bucket type turbines operated almost completely by impulse -- by the working fluid impacting the bucket. Today's axial flow turbines use blades mounted perpendicularly to a central hub, and rotate perpendicularly to the flow of gas through reaction -- or changing the direction of fluid flow. (Figure 2)

Tesla Turbine Design

Tesla turbines fall into the centrifugal category, but differ in the energy exchange mechanism. Conventional centrifugal turbines use blades to convert kinetic energy to shaft horsepower. Tesla's design uses the viscous effect of closely spaced disks, along with a number of small round washers to extract and convert the energy. (Figure 5)

Tesla stated that the round washers placed around the outer perimeter were absolutely necessary for start-up torque, and to give an advantage under highly loaded conditions. Understanding how the geometry of this outer periphery region interacts with the nozzle and the fluid passing through the nozzle is the key to disk turbine efficiency.

Phoenix Winglet Design

By replacing the round washers in Tesla's original design with thin-section winglets we are able to convert the kinetic energy of the gas using wing lift force (reaction) rather than the drag force of a round washer. As the fluid leaves the trailing edge of a winglet it continues to give up its energy through viscous effect -- if the winglet angle of attack is sufficiently small.

From the Labs

In our experiments we built up two disk packs which were identical in every way except for the geometric shapes placed between the disks. The "control" disk pack was built to Tesla's specs (using a spacing of 0.125 inch between the disks) and was used to compare experimental variations in the second disk pack. We used a small air compressor to charge a 20-30 gallon tank to 100 psi for each test cycle.

A frequency counter was used in conjunction with our custom built Hall-effect shaft rotation detector to compare runs. As long as every run begins with the same air pressure (100 psi), the disk pack configuration yielding the highest rpm is the most efficient.

While I won't go into much detail this month concerning test results, here is a brief summary:

a) Tesla configuration

average peak rpm: 

average peak rpm: 

Conclusions

Using a winglet in place of the outer periphery round washers of Tesla's design, we were able to gain approximately 30% higher efficiency. Based on other experimenter's test results with direct combustion and the Tesla configuration, we should expect our overall fuel to shaft efficiency to come in around 31% -- placing our design right between gas piston and diesel piston efficiencies.

Beyond Tesla

Next time we'll have photos of our test bed setup, and will discuss in more depth the aerodynamic differences between Tesla's original design and the Phoenix improvements. Till then, keep moving forward. We've got the sticks-in-the-mud on the run!

Ken Rieli

^ Top of Page