The Vision
To design and build a cold phase Wave Disk Generator for the purpose of gathering data to test numerical and computational models of wave engines under the consult of our client, Professor Akbari. The design proposed must be modular, easy to machine and assemble, and capable of being retrofitted. The design includes the integration of pressure transducers to gather dynamic pressure measurements to verify the accuracy of computational models of the pressure changes within the chambers of the Wave Disk Generator. Our efforts are to build upon past research in wave engine technology and utilize new theoretical approaches to computationally and physically verify designs for an in house prototype of a Wave Disk Generation utilizing the cold phase Wave Disk Generator for subsequent designs.
The Wave Disk Generator is a constant-volume combustion engine that utilizes compression of gas streams using expansion and expansion shock waves within rotating chambers to ingest and ignite a premixed fuel/air mixture more efficiently than existing combustion engines. The ignited mixture experiences further compression within a closed chamber to expel a high energy fluid that can provide work to a thermodynamic system. A cold phase Wave Disk Generator is a test engine supplied with compressed air instead of a high pressure air-fuel mixture for the purpose of gathering pressure readings of a fluid without ignition to later be compared to pressure readings of a fluid with ignition. It is a novel design for a compressor which has an end plate and has the potential to include combustion within its channels if it is upgraded for power generation. Shock wave compression can be utilized at high speeds of rotation where the fluid flow experiences sudden openings and closings of the inlet and exit inducing what we call expansion and expansion shock waves that affect fluid behavior. Our scope of the design does not include combustion and it is not necessary to measure the facilitating hammer shock resulting from a quickly closing channel. Turbomachinery numerical analysis proves that due to conservation of energy there will be a transitory pressure rise and temperature rise. By combining benefits of pre-compression in an internal combustion (IC) engines and complete gas expansion in turbine engines, existing heat engines can be redesigned to provide significant improvements in efficiency and drastically change the energy crisis. It has been well recognized that a machine operating on the principles of the Atkinson (aka Humphrey) cycle is more efficient than traditional IC and turbine engines.
At the onset of our project, we were asked to set up levels of accomplishment for our project group. They are as follows:
Levels of Success
Analytical Solution: Tier 1
Once these analyses were complete, in addition to material strength comparisons and procurement, a proof of design prototype was machined to show that such an device can rotate while enduring the loading of inlet and purging. This design took into account all the analytical results and should have shock wave propagation
We utilized 3-D printing for design and fabrication. The printed engine is void of other components of a gas turbine engine, and to size model of the desired model. The group took measurements of relevant thermodynamic properties at specific points and compared the experimental results to the analytical and computational results from the analyses performed. This proof of concept design can then be modified by subsequent models to improve on the machining, materials chosen, and other variables for greater efficiency.
Foundational Rotary Device: Tier 2
Once these analyses are complete, in addition to material strength comparisons and procurement, a proof of design prototype will be machined to show that such an device can rotate while enduring the loading of inlet and purging.
This design will take into account all the analytical results and should have rotation and static pressure rise; however, if our pressure transducer set up is incorrect we will not be able to see the waves pass by the localized region.
We will utilize Computer Aided Manufacturing software for design and fabrication. The machined engine will be void of other components of a gas turbine engine, and will be a scaled down model of the desired model. This prototype may not induce shock waves. The group will take measurements of relevant thermodynamic properties at specific points and compare the experimental results to the analytical and computational results from the analyses performed. This proof of concept design can then be modified by subsequent models to improve on the machining, materials chosen, and other variables for greater efficiency.
Due to our budget, we were not able to afford the typical alloy of choice for these types of machines: such as Inconel 718 used for proprietary turbomachinery blades. Therefore we limited our scope from a flame front shock wave engine to shock wave compression. Our initial plan was to buy Steel. We have accounted for the stresses of inlet pressure up to 24 atm gauge as well as rotating forces and found it to be well within tolerances. Our current design utilizes steel and 90psi gauge or 6 ATM inlet pressure with the parts in the design with a safety factor between 4 and 10. The only additional thing is the cyclical loading of shock waves. A housing was manufactured to these specifications with the goal of addressing the leakage and thermal expansion issues associated with the device. For this proposal, what is needed is to machine a wave disk engine using the optimal specifications obtained from our analytical research and budget constraints; taking into account radial leaning impellers. Our budget will be a key factor in testing the limitations of size on the wave disk engine.
Prototype utilizing specialized inlets, modular exit housing, at
varying channel angle: Tier 3
Beyond tier 2, we explored the effects of an increasing impeller angle and methods of attaching the inlet mass flow as well as the geometry and location relative to the inlet of the outer housing. This geometry and angle in incredibly important for the reason that it can have a profound effect on the shock wave creation and propagation. Too small of an angle could lead to wasted fluid, too high an angle would lead to a non-fully purged channel. This affects the increase of efficiency and leads to a more successful path to creating the WDG.
This solution, which is the goal of the client, is primarily based on the successes of the first two proposals and the integration of funding that would cover expenses of proper test and analysis rigging. We integrated an accurate control system to monitor wave propagation as well as combustion in the device. What were necessary, outside of a robust control system, were accurate sensors, transducers, and modular method of changing inlet and exit conditions.
Certain novel solutions to the problem of material strength could be explored with proof of concept thoroughly researched. For the future, depending on the foreseen utility, we would like to look into the Electro deposition machine of professor Kysar, and the laser shock peening process of professor Yao. The shock peening process would greatly help account for our material selection to compact the outer layer of the surface of out material. This compaction will help to increase case hardening which helps for durability during the combustion process. We did not test the combustion process with the current set up but have certain CAM layouts set up for when the future group will have the means to allocate for special material capable of combustion. [Laser shock peened rotor components for turbomachinery, Patent number: 5492447. Seetharamaiah Mannava et al] The EDM machine could be looked at as a novel approach to machining our design with better alloy, instead of high pressure die casting the design.
Proof of Concept, Compressor utilizing Shock Waves: Tier 4
Taking into account a proper inlet supply, control system, and data acquisition methodology, we were able to see pressure fluctuations thus proving the concept of pre-compression of fluid in the chamber. This concept allows for further research in the development of the device to allow for high efficiency combustion. Our target would be tier 3, with 2 designs of rotating chambers: one that has radial impellers, and the other with forward tangential leaning impellers.
Due to the high sampling rate and need for fine grain resolution of the data acquisition system, we used a powerful pressure transducer, that was limited by our lower level DAQ card. The minimum number of two transducers was used. One transducer was used to measure the pressure wave propagating in the chamber and one for the pressure at the exit. We realize that with unsteady flow we could investigate a novel way of normalizing the pressure at the exit through a duct work, the transducer for pressure exit would not have to be as accurate or as fast reacting as the one in the channel.
Some of the issues that was addressed when designing the prototypes included reducing the leakage within closed chambers. With the first of these issues, leakage will cause pressure loss and reduce the velocity of the fluid, creating discrepancies from the theoretical models. We dealt with the leakages by making interlocking designs and using silicone gel on the mated surfaces. Leakages, and pressure losses due to boundary layers most likely caused the measured pressure gains to be much smaller than predicted. It is not a compressor in and of itself so there should not be a pressure gain leaving to ambient it is to pre-compress the fluid for combustion which leads to more efficient burning and work produced within the chambers of the device.
To address the problem of if the pressure spike is not read: we went about eight design phases and remade our initial design from the ground up. We stressed the need to make it modular in order to make small changes to try and get the data to show. Beyond this: 1. We have a modular design that can be permutated, 2. We have fundamental turbomachinery analysis to take in new data sets 3. We have a CFD environment to move towards making a 3d design, or refining the 2d model to include tolerance 4. We have a control circuit model for the design 5. We have several stress analyses showing displacement of the impellers for each design made 6. Lastly we have a working prototype that is able to be retrofitted to adapt and address the failure points previously stated.
Results
After several months of consistent work, we were able to accomplish tier 3 with great assurance. From our data we believe we have accessed tier 4 as well, more data conditioning, and signal processing is needed to make sure. Either way, we are very proud of the outcome and the fact that we were able to succeed even in the face of adversity and many of those who thought we would not be even able to make a working prototype.
