Research

Flexible shaft coupling - US 3905208 A
This patent was attributed to Bridgestone Tire and is for a flexible shaft coupling. The body is made of elastomers with embedded steel reinforcements. The main novelty of this design is that the torque is quite low when the hub is initially put on the shaft and when it starts to spin, but after a short period of time the torque rapidly ramps up to the maximum torque output [1]. When the shaft rotates, the hub deforms radially inward which helps to increase the interference fit. A tighter interference fit produces more torque. More torque is produced the faster that the hub spins. This design is applicable to automobiles because you want a slow, controlled torque output. You do not want the maximum amount of torque almost instantaneously. Our design is not used for automobiles but we can leverage some of this patent`s ideas. We are constrained to use steel for our coupling in order to stay within Dresser-Rand`s specifications, so we cannot leverage the elastomer idea, but we can use the controlled torque ramp aspect. There are not many applications where an instantaneous ramp from zero to maximum torque is needed or even safe. Creating a safe power coupling is crucial because it will be operating at high rotational speeds and operators must be able to work around it. We may be able to use this patent to increase our torque output while using higher rotational speeds or we could use this to help reduce vibrations in our hub, which makes the design safer and more consistent.

Separation frequency analysis of interference fitted hollow shaft-hub connections by finite element method Using an interference fit is an easy and cheap way to transmit rotational motion and torque to a component. When the shaft-hub connection`s rotational speed increases, the interference stress between the shaft and hub can become zero at certain points. As such, this point cannot transfer any more torque and is known as separation frequency [2]. One of our team`s goals is to maximize the torque output for our shaft-hub. We can use this journal paper to try and find the separation frequency and determine where the weak locations are for torque maximization, and then design around those. The results of the paper show that shaft sizes of 100 - 500 mm produce the most prominent separation frequencies and should be avoided if at all possible. They used a finite element method to test their experiments and came to the conclusion that the traditional design method, based on Lame`s equation, produced unusable results for thick-walled interference fits. The diameter to interference fit ratio is a critical design component and the traditional model was not able to predict results if the ratio was less than 750. The finite element method will work if the proper mesh is used. The problem with using a high tolerance mesh is the amount of computing power needed to perform the analysis, which will be limited to us at UB. The entirety of this paper is based on the interference fit between a shaft and a gear hub but we can use the model and adapt it for our hub.

Global Shaft Coupling - US 6257985 B1
This patent is for shaft couplings, namely for shaft couplings that transmit torque between axially spaced shafts. The patent is on the geometry of the coupling components and the ways in which those components are assembled to form a coupling hub. The assembly of this coupling hub allows the ends of two axially spaced apart shafts to be closely aligned with one another [3]. The close alignment of the shafts will reduce vibrations when transmitting torque. In our project, one problem to solve is to prevent the shaft from digging into the coupling hub. Vibrations might be the reason why the shaft digs into the coupling hub. Therefore by using this design aspect of the patent, we might be able to solve that problem. It could also allow us to maximize the torque generated, a customer requirement given by Dresser-Rand. The ends of the two shafts can be spaced apart in ANSI, ISO, and DIN spacing increments in this design [3]. This is beneficial to connect shaft with different spacing standards without overhang or overextension [3]. Therefore, this design will eliminate the needs for additional components to prevent the overhang or overextension, making it easier to assemble and manufacture. We may be able to apply this aspect of the patent to reduce cost, another customer requirement given by Dresser-Rand. Overall, the two important advantages of this patent that may be applicable to our project are the closer alignment of shafts and the elimination of additional components to prevent overhang or overextension.

Analysis of a Shrink-Fit Failure on a Gear Hub/Shaft Assembly
This paper is on shrink-fit failure on a gear hub/shaft assembly. The strategy taken in this paper is calculating the stress as a function of the frictional force between the coupling hub and shaft, instead of the traditional method of calculating the stress as a function of applied torque. In this paper, two approaches were used to investigate shrink-fit failure, analytical approach based on Lames equation and finite element analysis using ABAQUS 6.4 software [4]. The purpose of this paper is to determine the cause of the failure, which is fretting damage [4]. The authors also estimated the maximum operating torque in which the micro-slip and the subsequent fretting damage were not significant. One of the main parts of our project is to maximize the torque output of our shaft. Therefore, we might be able to apply their result on the maximum operating torque for our project. In our project we have an interference fit instead of a shrink-fit. However, the two are not that much different so we might be able to use both the analytical approach as well as the finite element analysis for our project. In our project, Dresser-Rand wants us to use ANSYS software to perform a finite element analysis. We might be able to apply a similar model in the ANSYS software to the one they used in their ABAQUS software. According to this paper, the analytical approach they took gave a more accurate result as opposed to the finite element analysis, so we should also use that for a more accurate result [4]. Overall, we can investigate our coupling hub/shaft assembly in a similar way used in this paper.

Rotational torsion fatigue failure of an engine driven fuel pump coupling
This paper [5], published in the Fatigue & Fracture of Engineering Materials & Structures (FFEMS) journal, points to a study conducted by the Defense Technology Agency for the Royal New Zealand Air Force. In this study an aircraft engine used by the air force suffered multiple coupling failures to the shaft attached to the engine driven fuel pump. The investigatory team was tasked with finding the cause of the failure. Having analyzed the point of failure, it was determined that reverse torsion loading in addition to vibration loads may have caused fatigue failure of the coupling. To confirm their hypothesis an aircraft was run on the ground through various power settings and accelerometers attached to engine record the vibration response of the engine. These measurements were then used to calculate inertial loads and from this fatigue analysis was performed to test for theoretical failure conditions. It was also found that load reversals from the changing of engine power added to the torsion stresses experienced by the coupling. In his conclusion the author stated that engine service life, its usage and time between maintenance periods all played a role in the failure of the component. The reduction of vibration by increasing the frequency of maintenance cycles seems to be the primary solution suggested by the author. However despite solving the problem the component remains unchanged leading to possible failure if the engine is not regularly maintained; in essence the component life was increased by increasing maintenance cost. Fatigue analysis similar to this could be used to calculate the fatigue life of the coupling designed and evaluate ways to improve that life.

Flexible Coupling - US 1983094 A
This patent [6], presents a patent for a coupling, one that may present a solution to the problem described in the previous paper. Vibration encountered in the previous paper resulted as a consequence of a well-used engine running. The shaft connecting the fuel pump to the engine was solid and rigid. This patent presents a coupling which is flexible and is called, by definition, the flexible coupling. The flexible coupling works by transmitting toque through a composite interface of rigid metal and rubber like material. The flexible material transfers small amounts of torque, while the metal transmits higher torques. This combination of material allows for the shafts to be offset to each other. A vibration produces an inertial load causing the shaft to move with respect to the coupling. The use of elastic material in the coupling causes a force due to vibration to be isolated from the metal part of the coupling. This enables the vibration and the respective force generated to be isolated preventing metal fatigue altogether. A coupling with such an interface could last longer and would not need to be replaced as often. This would lower assembly cost and make for less frequent removal and inspection periods. Adding similar features to our couplings could reduce the frequency of disassembly and assembly leading to less damage to the shafts. The use of elastic material could also lower the force required to expand the hub, something that our customers are looking for. However the use of hot oil, used in the current design, as a hydraulic fluid may limit the use of material that can be safely used.

Means for directly coupling main motor shaft to working machine - US 6772660 B2
While researching relevant patents for our Dresser-Rand Power Coupling project, we came across a method for directly coupling a shaft to a working machine. This invention provides a direct means for torque transmission which is achieved by a premade installation cavity for the shaft allowing for seamless engagement as opposed to a coupling which would connect the shaft to the working machine [7]. This design for a direct connection eliminates the reliance of skilled workers or further machining for precision alignment between the shaft and the coupling hub. It also reduces high production costs by cutting wasted labor time for assembly, therefore improving marketability and overall profit. By eliminating the traditional two piece assembly the effects of vibration on the shaft are reduced allowing for smoother operation and less wear on the system as well as increased efficiency and a prolonged life cycle. This idea would benefit our current design by removing the intense hydraulic pressure required to expand the hub and thus preventing the damage from the hub digging into the corresponding shaft. The only issue is the potential maximum torsional output from this design and if it can create the necessary power required for our customer. Also due to the environmental conditions and wear from operation this one piece, direct connection, might not be able to withstand the effects. Further analysis of this design would be essential to determine whether or not it could be adequate enough for our required application.

Numerical and experimental investigation of the torsional stiffness of flexible disc couplings
Couplings are common to mechanical engineering in that they are used to connect two rotating pieces of machinery. The disc coupling is one form of coupling a shaft to a working machine. It has gained popularity due to its flexibility which reduces misalignment issues by accommodating discrepancies in axial, radial, and angular directions [8]. As opposed to rigid connections, whose stiffness can cause damaging vibrational stresses, disc couplings reduce vibrational effects and can increase length of efficient operation due to their flexibility. This concept could benefit our design project considering we are attempting to develop an efficient coupling connecting a shaft to a working machine. The stresses applied to the shaft and hub during assembly, operation and removal create an angular displacement which hinders the performance and maximum possible torsional output. In our case, those stresses also cause the hub to dig into and damage the shaft as well. By using a more flexible coupling we would be able to reduce the hydraulic pressure required to expand the hub for shaft insertion and thus eliminate any digging due to deformation of the hub. This would also help to reduce any vibrational effects that may have compromised the functionality, torsional output, forcing fatigue, and consequent removal of the coupling for repairs to the system. The journal provides experimental data and analysis of the relationship between torque and angular displacement by examining the corresponding results of varied coupling stiffness. The data shows a correlation between torsional stiffness and displacement. By reducing the coupling stiffness we can reduce any misalignment and increase torque with less wear to the system.

Flexible coupling with end stress relief structure - US 5611732 A
This patent was granted to TB Wood`s Incorporated and serves the purpose of transmitting torque between a drive shaft and the driven shaft. Although the shafts are aligned on the same axis, there will be some shaft play which is where this patent evolves from. In order to accommodate the shaft play a flexible coupling was innovated utilizing an elastomer as the coupling. Because the coupling was designed with two halves, this causes stress concentrations because the torque applied causes twist that is not applied evenly. The metal shoe is implemented to fix the stress concentrations. The goal of our project is to be able to remove a shaft from a coupling that was initially an interference fit, without scratching the shaft or coupling surface. The problem is that it is not known when the actual scratching happens, whether it be during insertion, under load, or upon removal, so we have to treat all three as the problem. There will be shaft misalignment under load, and because this could be a potential cause of the scratching, making our coupling flexible could solve the problem. Even using a different material on the inside of our steel coupling like a sleeve, such as an elastomer could prevent scratching and even if it was scratched, allow us to just replace the cheap elastomer sleeve and not the entire steel coupling. An elastomer sleeve could also help reduce vibrations experienced during operation.

Pay Close Attention to Coupling Alignment
Shaft misalignment is the main cause of vibration which can lead to coupling failure, shaft failure, and oil leaks. Our Dresser-Rand power coupling uses oil and high pressure seals so it is important to pay attention to the orientation of the shafts under load. Although flexible couplings can correct some shaft misalignment, there will still be stresses due to the misalignment, so it is important to initially have the shafts line up as concentric as possible. Two types of misalignment can happen; axial misalignment and angular misalignment. Our coupling can be exposed to each and they can both be the reason why we have some scratching on the coupling and shaft after they are separated from each other. Any angular misalignment would cause one side of the shaft to dig into the coupling and any axial misalignment would also cause a large vibration in our case, which could definitely result in some scratching of the coupling. The vibration could also cause the o-rings which seal the oil in the coupling to fail, allowing leakage, ultimately causing scratching during installation and removal of the shaft. The best way to prevent shaft misalignment is to correct it from the start. Some methods involve using a straight-edge, calipers, computer programs, and laser beam alignment. The straight edge and calipers are not accurate enough for our application as we have a very high torque at a fairly high rotational speed. Laser beam alignment is the most practical for us as it provides the most accurate alignment. Shaft alignment is very important, and it is better to deal with it at the start of the assembly, not after it is already together.

References
[1] Yoshinori, M., Fujio, O., and Hiroichi, O., 1975, "Flexible Shaft Coupling," U.S. Patent 3905208.
[2] Volkan, K., 2011, "Separation Frequency Analysis of Interference Fitted Hollow Shaft-Hub Connections by Finite Element Method," Advances in Engineering Software, 42(9), pp. 644-648.
[3] Ward, E., and Buckbee, M., 2001, "Global Shaft Coupling," U.S. Patent 6257985.
[4] Truman, C., and Booker, J., 2007, "Analysis of a Shrink-Fit Failure on a Gear Hub/Shaft Assembly," Engineering Failure Analysis. 14, pp. 557-572.
[5] Withy, B., James, A. and Williams, J. 2012," Rotational torsion fatigue failure of an engine driven fuel pump coupling". Fatigue & Fracture of Engineering Materials & Structures, 35: pp. 37-44.
[6] Neher, J. George, 1934, "Flexible Coupling," U.S.Patent 1983094A.
[7] Tsai, C. H., Hwang, S. J., and Chen, Y. T., 2004, "Means for directly coupling main motor shaft to working machine", U.S. Patent 6772660 B2.
[8] Zhao, B., Zhao, Y., Feng, J., and Peng, X., 2016, "Numerical and experimental investigation of the torsional stiffness of flexible disc couplings", International Journal of Mechanical Sciences, 114, pp. 207-216.
[9] Tirumalai, S., 1975, “Flexible Coupling with End Stress Relief Structure,” U.S. Patent US 5611732 A
[10] 33 Metalproducing, 2001, “Pay Close Attention to Coupling Alignment.” pp. 30