The field of the invention is the connection between a shaft and a coupling and more particularly in the context of downhole equipment where the shaft is splined and the splines are configured for stress distribution.
Equipment designed to fit in bore holes is, by nature, long and slim. Torsional loading of the shafts is often a limiting factor in the maximum power developed. Shaft connections are critical weak points in the shaft designs. The preferred choice for shaft connection in an Electrical Submersible Pump (ESP) is a spline, either a multi tooth involute, or six tooth square spline. Both have proved to be quick and reliable for field assembly of the equipment. Shaft material strength requirements for high power units can exceed 200,000 PSI yield. This is reaching the limits of readily available materials. To achieve these strength levels, elongation (ductility) has to be sacrificed. The selected materials are therefore more brittle and more susceptible to fatigue in the areas of high stress concentration.
The object of this invention is to reduce the high concentration of stress by making it possible to more uniformly distribute the forces involved along the spline. The problem of the spline joint is, when under load, the coupling which is stiffer will twist much less than the shaft. This is due to the difference in diameter. The angle of twist per unit length in a shaft varies directly as the applied torque, inversely as the modulus of shear and the forth power of the diameter. The coupling, even though it is a hollow cylinder, has to be extremely thin to match the angle of twist of the shaft. This is impractical because the coupling thickness is necessary to counter the radial forces that are attempting to expand (explode) the coupling. For example, if the shaft is 1″ diameter and the coupling is 1.0″ ID and 1.5″ OD, the coupling is five times stiffer than the shaft. Under the same torque load, it will only have ⅕ of the twist as a shaft of the same length.
Examining failed shafts shows that almost the entire load is distributed over a short area where the end of the coupling engages the shaft spline. The length of this “work area” can only expand as the twist of the coupling allows more contact with the shaft and the local mashing (yielding) of the spline tooth itself. The solution to this situation in this invention is to shape the spline so that the initial contact area begins at the end of the shaft in the middle of the coupling. As the shaft twists in relation to the coupling, the contact area increases, the stress is spread over an increasing area, reducing the stress concentration at the initial contact location.
The coupling is formed by a pull broach, and it would be an extremely difficult operation to create a coupling spline that was wider at the start (mouth) of the coupling and taper as it progressed inward. The shaft spline is more easily modified. For a square tooth spline on which the sides of each tooth are parallel, the cutter would have to progress down the shaft with a slight twist angle to the axis of the shaft. It would also be necessary that the mating coupling spline tooth be cut wide enough to accommodate this twisted spline. A simpler arrangement could be made for the involute. These splines are generally made on a hobbing machine. The shaft is rotated and a rotating cutter, timed to cut the spline teeth progresses along the shaft from the end to the maximum design length of the spline. If the cutter plunges into the shaft material as it progress along the spline it will result in a tooth width that is wider at the end of the shaft and tapers as it moves into the spline. This will force the coupling to make first contact in the depth of the coupling and the contact area increases in the direction of the mouth of the coupling as the shaft material twists in relation to the coupling. The plunging cutter method will not work on a true square tooth form, but can be applied if the form is modified so that the shaft tooth tapers towards the OD of the shaft.
The present invention addresses stress concentrations that occur in a splined shaft that is involved with the transfer of torque with a coupling. This concentration is caused by the comparative rigidity of the female spline coupling which encompasses the male spline shaft. This difference in magnitude is sufficient to allow an assumption in the design that the female spline is rigid and all the torsional elastic displacement is concentrated in the shaft in a very narrow band of material at the intersection of the mouth of the coupling and shaft. As the mechanical limits of the equipment are pushed, stronger shaft materials are the option of choice. These metals obtain strength at the sacrifice of ductility and a heightened sensitivity to crack propagation reduction in fatigue resistance.
To lessen the burden on the shaft it is necessary to increase the area of the torsional contact beyond the narrow band. Increasing the compliance of the coupling by reducing its diameter is impractical because it increases the radial stress in the coupling from the radial component of the torque developed by the spline tooth. The coupling could fail from expansion (explosion). The possible solution is to have the axial tooth contact line designed so that as the torsion deforms the shaft elastically, the contact line becomes longer. This can be accomplished by changing form of the involute or square tooth so that it slightly spirals in the opposite direction of the torque loading in anticipation of the a elastic strain that occurs with loading. This is just marginally practical. Basically a large amount of tedious work is required for the benefit achieved. A simpler method of the present invention is to sacrifice a very small amount of shaft strength. This method requires that tooth form be developed so the tooth cutting hob plunges slightly towards the center of the shaft as the cut runs from the end of the shaft to its maximum length. The result is a “tapered tooth” wider at the end of the shaft which will first make contact with the coupling near its center and extends its contact line as the shaft deformation increases.
The following US Patents deal with the issue of splined shafts and couplings and the general design considerations that are involved in these designs but do not touch on the issue of stress distribution in such designs as addressed by the present invention: U.S. Pat. Nos. 3,073,134; 4,133,516; 4,331,006; 6,056,511; 6,120,261; 6,126,416; 6,332,841 and 6,709,234.
Those skilled in the art will gain a better understanding of the invention from the description of the preferred embodiment and associated drawings that appear below with an understanding that the full scope of the invention is given by the appended claims.
A connection between a shaft and a coupling that preferably involves involute shaped splines is designed to reduce stress from torque applied to the shaft by a drive motor. To avoid stress concentration at the mouth of the coupling on startup, the shaft splines are made to taper from the shaft end down to the spline end. As a result, the initial contact with the coupling splines occurs within the coupling and removed from it mouth. The shaft is allowed to elastically twist in response to the applied torque as the contact area increases with shaft twisting and the contact line moves toward the mouth of the coupling. Some shaft elastic twist occurs without adding stress to the shaft splines until coupling contact is made.
a is an enlargement of the dashed circle in
a is an enlargement of the dashed circle in
The situation addressed by the present invention is shown in
As a result when torque is applied to shaft 32 the initial stress is well within the coupling 60 as the contact of splines 40 and 56 starts there and as the shaft 32 twists in reaction to the applied torque the gap 54 closes in the direction toward the mouth 58 of the coupling 60. As the spline contact area increases due to elastic twist of the shaft 32 the contacted stress is reduced as the torque is transmitted over a progressively increasing contact area between the mating splines 40 and 56. In contrast to what happens in the
Those skilled in the art can now appreciate that cutting the splines on a shaft or the mating coupling at an angle to the central axis allows minimization of stress at the coupling mouth where in the past failures from torque induced stress on startup were most likely to occur. Instead, contact between the mating splines initiates within the coupling and spreads as the driven shaft twists and the spline contact area is increased. By the time the gap closes at the coupling mouth the stress level on the driven shaft splines is reduced. The applied toque induces an elastic twist in the driven shaft that places progressively more splines in a meshing relation while sparing the shaft spline at the mouth of the coupling from being immediately stressed as in the past with the
The above description is illustrative of the preferred embodiment and many modifications may be made by those skilled in the art without departing from the invention whose scope is to be determined from the literal and equivalent scope of the claims below.
Number | Name | Date | Kind |
---|---|---|---|
1381805 | Crewdson | Jun 1921 | A |
1803995 | Chilton | May 1931 | A |
2038554 | Edgar | Apr 1936 | A |
2297390 | Burger | Sep 1942 | A |
3073134 | Mann | Jan 1963 | A |
3222772 | Leyner | Dec 1965 | A |
3360961 | Steiner | Jan 1968 | A |
3742656 | Amos | Jul 1973 | A |
3865500 | Newell | Feb 1975 | A |
4115022 | Orain | Sep 1978 | A |
4133516 | Jurgens | Jan 1979 | A |
4331006 | Bishop | May 1982 | A |
5533825 | Stone | Jul 1996 | A |
5660494 | Schwarzler et al. | Aug 1997 | A |
6050346 | Hipp | Apr 2000 | A |
6056051 | Coronado | May 2000 | A |
6056511 | Kennedy et al. | May 2000 | A |
6120261 | Al-Towailib | Sep 2000 | A |
6126416 | Lee | Oct 2000 | A |
6332841 | Secord | Dec 2001 | B1 |
6547479 | Dowling, Jr. et al. | Apr 2003 | B2 |
6604885 | Neuner | Aug 2003 | B1 |
6709234 | Gilbert et al. | Mar 2004 | B2 |
6752560 | Wilson et al. | Jun 2004 | B2 |
7223074 | Rydberg et al. | May 2007 | B2 |
7288029 | Lyon et al. | Oct 2007 | B1 |
7478969 | Kozlowski et al. | Jan 2009 | B2 |
20090291003 | Brown et al. | Nov 2009 | A1 |
Entry |
---|
Southard, Robert, “New Assembly Drills Without Reactive Torque”, SPE/IADC 105000, Feb. 2007, 1-7. |
Number | Date | Country | |
---|---|---|---|
20100303537 A1 | Dec 2010 | US |