US2349146A - Engine starting gearing - Google Patents

Description

May 16, 1944. I J. B. DICKSON 2,349,146

ENGINE STARTING GEARING Filed May 10, 1943 s Sheets-Sheet 1 INVENTOR E a WMZMW I [2' fi/m ATTORNEYS- May 16, 1944.

J. B. DICKSON ENGINE STARTING GEARING 5 Sheets-Sheet 2 Filed May 10, 1943 1 INVENTOR p k70 i)? Elvira/ mhww m ATTORN EYS May 16, 1944. .1. B. DICKSON 2,349,146

ENGINE STARTING GEARING Filed May 10, 1945 a Sheets-Shani INVENTOR J56)? ,3, Eel 5am ATTORN EYS.

latente cl May 16, 1944 "ENGINE STARTING GEARIN G John B. Dickson, Highland Park, Mich, assignor to Chrysler Corporation, Highland Park, Mich a corporation of Delaware Application May 10, 1943, SerialNo. 486,327

6 Claims.

This invention relates to an engine starting gearing, and it has particular reference to the provision of means in a threaded runner member and stop abutment for assuring the free movement of the runner away from the stop and the elimination of binding therebetween.

In a known and extensively used form of starting drive for internal combustion engines, an electric motorrapidlyaccelerates a screw threaded shaft carrying an internally threaded pinion and a stopnut or end abutment secured to the shaft. The inertia. of the pinion causes it to run along the shaft screw and into -engagement with the stop nut, whereupon the pinion and shaft revolve in unison. As the pinion reaches. the limit of its travel, it also meshes with a gear formed on the engine flywheel, or otherwise connected to the crankshaft, and thus spins theengine, enabling it to start.

As soon as the'engine begins firing, it revolves at a relatively high speed, and drives the pinion faster than the motor shaft. The reversal of forces accordingly is supposed to make the pinion travel in the opposite direction along the shaft screw,and to disengage from the engine gear. In order to facilitate this intended movement of the pinion,'the screw ismade with a fairly long 1ead-multiple threads being employed to conserve space. Such construction embodies what is sometimes called a free running connection-if the screw shaft and pinion are held in a. vertical position and one memberonly is held against rotation, the other will rotate and travel lineally under'its own weight. Generally speaking, such free running connection has been found efficacious and reasonably devoid-of service failures 'when'used in starting gearings, and

the principle has also been utilized in other mechanical movements.

Sometime ago, however, it was noted that excessive failures were encountered when applying such gearing to'the internal combustion engines of military tanks. "While the-pinion would engage in the manner described, it would fail to disengage when the engine started. As a consequence, the motor armaturewas driven at. such a high' speed as 'to cause'it to burst-from the centrifugal forces developed. Inspection ofsuch installationsshowed an absence of any'known cause for failure, such as excessive dirt,'shaftr5'0 bending, or the like.

I have now discovered What I believe to be the nature of the phenomenon'acausing the freely running pini'on member tot remain "in engageover devised a relatively simple means of eliminating the cause-of such fai1ureswhether occurring in tank installations or in any other situation where'the effect might be encountered.

.5 A further discussion of the subject will be facilitated by referring to the accompanying'drawings, wherein:

Fig. 1 isa plan View showing a starting vmotor having a pinion adapted to engage intermit- '=.1.0 tently a driving gear for an internal combustion engine.

Fig. 2 is a section taken (along the line .2-2 of Fig. 1, and drawn to an enlarged scale to show a structure embodying the principles .of

,115 this invention.

Fig. 3 is a fragmentary section showinga pinion and stop nut in engagement.

Fig. 4 is a view similar to Fig. 3 but showing the nut and pinion in spaced relation.

.20 Fig.5 is a perspective'of the form of stop nut shown in Figs."2,3, and 4.

Fig. 6 is an elevation of a prior art assembly showing how a binding effect is developed in a 'free running construction, and how it -may be 1 25 eliminated.

Fig. '7 is an elevation showing the determination of some of the angles of the nut illustrated in Fig. 5.

Fig. 8 is-a .perspectiveshowing another form :30 of stop nut.

4 0 the pinion 23 vmoves into mesh with the gear 24 and also abuts a stop nut '26 secured to the -.driven motor shaft. As thefiywheel'25 isbrought :up to speed, the engine cylinders beginfiring r and thus cause the'flywheel to revolve at ahigher speed than that imparted-bythe motor .drive. -When this occurs, the pinion "i23'is intended .to .move out of engagement with the gear 24, and accordingly return to the position shown .in Fig. 1.

Some of theforegoingparts are more clearly illustratediin. Fig. .2 from which-.it will. appear that the pinion 23 ismounted upon a motor driven shaft 21 formed .with a multipleithread engaging complementary threads within the bore ment with the -engine gear, and I have moreminor thepinion. .I-he .nutl26- is securely pinned to the driven shaft 21 by means of a connection 28. It will thus be understood that when the shaft 21 is accelerated by the motor 2|, the pinion 23 runs along the shaft threads until it both engages the gear 24, and abuts the nut 26. Since the pinion can no longer travel along the shaft it must now rotate and, therefore, cause the gear 24 and connected engine parts to revolve.

Figs. 2 and 3 show respectively the pinion 23 abutting a stop member 29 and the nut 26, at the extreme limits of its travel. pinion is shown at an intermediate position, which may here be considered as either approaching toward or receding from the end abutment 2B. These figures also show that the screw 3| formed In Fig. 4 the on the shaft 21 is multiple threaded and in-fact includes the three threads, 3|a, 3H), and,3lc. As thus far expressly described, the structure is one heretofore known and usedthe specific differences over the prior art being in the formation of the abutting faces of the pinion 23 and the nut 26, and the mode of operation upon disengagement. Heretofore, these abutting surfaces have been transversely perpendicular or normal to the axis of the shaft 21, as shown in Fig. 6. It will be seen in the remaining figures that they are illustrated as being formed with angularly disposed serrations which, according to Fig. 3, may match and engage perfectly, or, according to Figures 8 to 10, may engage but not over as great an area as in Fig. 3.

In order to explain fully the significance and effect of these serrated faces, reference may be made initially to Fig. 6. Let the points A and B be two points in the same radial plane intersected by any one thread, such as the thread am, wrapped on a cylinder Whose effective diameter is Dr. Then, in one rotation of the screw, a nonrotating nut at the point B will move to the left a distance equal to 13; which is usually called the lead of the thread. The angle which the thread helix makes with a transverse normal plane through the point A or B is then given by the equation:

AB tan 0 From this, the value of the angle 0 may be readily calculated, and in IE. 6 it has been presupposed that the distance AB is exactly one inch, while the thread 3| has an effective diameter of 1.125 inchesan actual case. From this the value of the angle 0 may be readily calculated as bein approximately equal to 15-48. This lead or helix angle for the Acme type of thread illustrated is normally free running, and adequate to throw the pinion into mesh with the gear '24 when the motor 2| is accelerated.

An explanation will now be given of the apparent paradox of the prior art pinion 23a failing to disengage from the stop abutment Z ta when the lineal thrust on the running member is reversed as by the starting of the engine. At such time, the normal end face 32 of the pinion 23a is in contact with the parallel end face 33 of the abutment 26a, whose outside diameter, (Dn), is taken to be 1.6875 inches. The surface 33 therefore presses against the surface 32, causing the internal threads of the pinion 23a to be forced against the screw threads 3| by an amount proportioned to the area in contact (or the stop diameter), the direction in which the pressure is applied, and the resistance of the engine to starting. While an increase in the total force thrusting the running member to the right should facilitate separation of the surfaces 32 and 33- the screw 3| having a large helix angleit actually may fail to do so, as the above described experiences reveal.

The pressure between the surface 32 and 33 is, of course, created by running the pinion to the left, and the condition for free running is, that no amount of end pressure, within the limits of rigidity of the parts, will cause the screw 3| and the pinion threads to bind on each other. However, as the distance from the shaft axis to the outer diameter of the abutment increases, it will be seen that the trace of a screw having the same lead as that of the threads 3|a, 3|b, and 3|c becomes progressively flatter, and the pressure angle between the thread and the nut surfaces is acco rdingly diminished. Thus, if a screw of lead AB be traced on the diameter Du, then its helix angle, qs, is such that It follows that the angle is less than the angle 0 for any value of Dn greater than Dr. Numerically, in Fig. 6 the angle 5 is reduced to 10-41, while if the nut diameter is increased to two inches, the angle is reduced to 93'.

It accordingly appears that, having in mind the radius of the contacting surfaces 32 and 33, then a binding force can be developed between the screw and pinion threads, of such magnitude as to wedge the parts together. In effect, the areas of the normal surfaces 32 and 33 subtract from the freedom of movement imparted by the pitch of the screw, or diminish its effective helix angle as the areas are radially increased. This concept was tested by replacing the nut 25a with one two inches in diameter, whereupon binding, and disengagement failures were even more pronounced. Reduction of the diameter of the abutment is not indicated, however, because of loss of strength.

In order to put this concept into practical form, and to overcome the wedging effect caused by the diminution of the effective pressure angle, the I surfaces 32 and 33 are accordingly so rearranged that they are inclined to a normal transverse plane to intersect the same, rather than be parallel thereto. Thus, in Fig. 6, let the line 34 indicate the trace of a transverse plane, perpendicular to the plane of the paper, and normal to the axis of the shaft 21. The helix angle of the screw 3| is then 0, or the angle between the plane 34 and a plane or line 35 tangent to the trace of the screw, while the angle is the helix angle of a screw of the same lead, generated on the outer diameter of the abutment 36a, with reference to the normal surfaces 33 or 32, and Whose trace is tangent to the reference line 36. If the surfaces 32 and 33 be turned away from the normal through some such angle as sov as to lie in a new reference plane 31, then the inclination of the helix angle to the outer diameter is no longer but is When this is done, the pressure or wedging angle is increased, and binding is curtailed. If the assumed angle 4; is made equal to the calculated angle then the effective pressure angle at. the outer diameter is also 0, or an angle which is free running. If the angle p' is made greater than there is even less than normal probability of any jamming from dirt or similar causes, while if the angle lies between zero and some advantage is gained, but not the fullest amount. .Similarly, if the angle be made negative with respect to 6, then the effective wedging or pressure angle will not be increased, but will actually be more acute, and therefore more likely to bind than the plane surfaces 32 and 33. What is wanted is an angle, such as that marked-(+') so as to preserve the thrust component inducing the runner member 23a to move away from the abutment 25a under a force acting toward the right, and, at the same time, preserve the freedom of travel from the right into holding engagement with the stop.

It will now be apparent, in View of the permissible limits for the angle that many specific forms and shapes maybe adopted to apply the principles of the invention in a practical way, and two cases will be discussed as sufficiently illustrative. In Figs. 2-, 3, 4, 5 and 7 the pinion runner 23 and the abutment nut are formed with siX meshing teeth or serrations, while in Figs. 8 to weight are employed, with some other changes hereinafter pointed out.

In the nut shown in Fig. 5, which is adapted to be paired with the pinion 23, there are six teeth T whose flanks M and 4 2 meet in a crest 43, and wherein adjacent teeth join in valleys 44. The pinion 23 is "formed with complementary teeth T whose flanks 65 and 45 are adapted to mate with the flanks 52' and 4|, the pressure being developed by engagement of the flanks 45 and 42. I

Assuming that the nut 26 has a facial diameter, Dn=2 inches, and that the screw of Fig. 6 is to be retained, then the value of 0, as projected onto Dn, is, as stated above, 4 =93.. crests and valleys 43 and M are laid out on radial lines, then the depth of each tooth can be determined from the geometry of the figure. For the purposes of ease of manufacture and symmetry, the flanks ti! and 12 are made of the same size the crests 53 being located midway between the valleys 44. These factors will therefore give a relationship such that the contacting faces cannot create a wedging or binding force subtract- If the ing from the freedom inherent in forming the V screw 3| with a helix angle of 15? or 16. Moreover, since there is a definite relationship in terms of lead, between the pitch of the screw 3| on the shaft 21 and the angle on the stop nut 26, a forward motion of the pinion 23 toward the nut 25 will cause the crests of the teeth T to clear the crests i3, and come to rest in the valleys 44 with the surfaces Q5 and :32 in full contact (normal machining tolerances neglected) There will, therefore, be a large amount of contact area between the end faces of the mating parts, as shown in Fig. 3.

Referring particularly to Fig. '7, it will be seen that, while the flanks of the teeth T and T are planar, and while these boundary crests and valleys lie radially of the end faces of the runner 23 and abutment 26, such boundaries are not normal to the axis of the shaft 21, but all have a common locus in such axis. The crests and valleys on the pinion and stop therefore slope in opposite directions. The reason for this will be more clearly understood by considering each pair of contacting flanks 1-5 and 12 to be sectors of the original surfaces normally disposed as indicated in Fig. 6. If these be so turned as to develop the angle 5 at the outer circumference, the boundary radii will thus be inclined toward the focal point or center in the shaft axis. Otherwise, these sectors would be twisted upon themselves like so many vanes of a pinwheel, and contact could not. therefore, be obtained except along a line or over a very narrow zone. Proper inclination of the crests and valleys therefore insures complete surface contact over the entire flank. Moreover, it has been pointed out that the angle may decrease as the diameter increases, and it can now be visualized that the I inclination of the sector or flank boundaries at a proper angle may automatically compensate for the change in angle within increasing radius of the abutment.

The flanks are therefore cut with their boundaries at an angle to the central locus, as well as at an angle e with reference to the helix angle of the screw. The determination of the taper angle is again a matter of shop mathematics preliminary to setting up the milling cutter, or other forming tool. For the dimensions given as illustrative, the angle is 2-26'. With this modification, it will be seen that the helix angle 6, when projected to any diameter on the contacting surfaces, preserves an equivalent non-wedging pressure angle The case just described is for full circumferential contact or mating of the teeth, wherein the selected angle was equal to c, or 0 as projected to the stop periphery. Such a construction was found to give excellent results under the same service conditions causing the failures first referred tothe pinions engaging and disengaging indefinitely.

Under some field conditions, however, dust and gum will creep into the space between the screw 3| and the threads of the pinion, thus causing an extraneous resistance to the free movement of the pinion along the motor shaft. This can be remedied by cleaning if the pinion fails to engage, but if engagement once takes place, failure from dirt inclusions is productive of a ruined motor and therefore the stalling of the entire power plant.

In the nut and pinion assembly of Figs. 8 to 10, this possibility has been secured against by increasing the added angle 75 above that employed in the first described form. The angle is selected as being approximately 17, and the number of teeth are increased to eight. To preserve this angle, the draft or taper between the engaging flanks is increased to 5-8'. Since the angle 5' is now greater than the angle 0 as projected, there is no wedging action between the surfaces 52 and 55 formed on the stop 26b and the pinion 23b. If the pinion is therefore sufficientl free on the screw to be thrown into meshing engage.- ment, the starting of the engine will cause disengagement before the motor is over-speeded. It will be noted, in Fig. 10, that the pinion and stop nut teeth do not engage with their valleys and crests in alignment, but that the flanks 55 run past the valleys 54 and engage only about half of the area of the flanks 52. It will thus be seen that the subsequent movement of the pinion 23b to the right is not induced by any camming action, as might have been assumed from a superficial consideration of the substantially continuous contact illustrated in Fig. 3. This is now apparent since, in Fig. 10, the surfaces 55 and 56 can never come into contact with the flanks 5i, due to the necessity of clearance. The embodiments of the invention should not, therefore, be confused with conventional types of saw tooth clutches.

While the invention has been described with reference to two specific forms thereof, and one service application where it has been found to have great utility, it will be understood by those skilled in the art that the principles may be embodied in other forms and utilized in other relationships. It is also to be understood that while I have stated that rationale which appears best to conform to the observed difiiculties and improved results, the observations and improvements are as stated, irrespective of the ultimate accuracy of any matters of theory. Accordingly, it is me tended that the foregoing description be regarded as illustrative, and that the scope of the invention determined from the following claims.

I claim:

1. In the combination of a screw having a stop abutment and an internally threaded runner member mounted on and adapted to travel lineally of the screw into and out of engagement with the stop abutment, a stop abutment and a runner member formed with complementary surfaces engagingv each other with sliding Contact as the runner member advances against the abutment, said surfaces forming an angle with a tangent to the trace of the screw greater than the helix angle of the screw.

.2. In the combination of a screw having a stop abutment and an internally threaded runner member mounted on and adapted to travel lineally of the screw into and out of engagement with the stop abutment, a stop abutment and a runner member formed with complementary surfaces engaging each other with sliding contact as the runner member advances against the abutment, said surfaces forming an angle with a tangent to the trace of the screw, said angle, at the outer diameter of the contacting surfaces, being greater than the helix angle of the screw by an amount such that the ratio of its tangent to the tangent of the helix angle is equal to or greater than the ratio of the screw diameter to the outer diameter.

3. In the combination of a screw having a stop abutment and an internally threaded runner member mounted on and adapted to travel lineally of the screw into and out of engagement with the stop abutment, a stop abutment and a runner member formed with complementary surfaces engaging each other with sliding contact as the runner member advances against the abutment, said surfaces forming an angle with a tangent to the trace of the screw greater than the helix of the screw, the boundaries of said surfaces forming a second angle with a normal to the screw axis intersecting said tangent, the vertex of said second angle lying substantially in said axis.

4. In the combination of a screw having a stop abutment and an internally threaded runner member mounted on and adapted to travel lineally of the screw into and out of engagement with the stop abutment, a stop abutment and a runner member each formed with a plurality of complementary teeth adapted to engage each other when the runner advances against the abutment, said teeth having mutually engaging flanks bounded by crests and valleys lying along substantially radial lines having a locus in the axis of the screw, said engaging flanks forming an angle with a tangent to the trace of the screw greater than the helix angle of the screw.

5. In the combination of claim 4, engaging flanks on the teeth of the runner member and stop abutment whose inclinations to the screw axis and a tangent to the screw are such as to satisfy the proportion:

where is the inclination of the flanks to the tangent to the screw, 0 is the helix angle of the trace of the screw, Dt is the effective diameter of the screw, and Dn is the diameter of any circle lying between the inner and outer surfaces of th flanks whose center is in the screw axis. 6. In the combination of a screw having a stop abutment and an internally threaded runner member mounted on and adapted to travel lineally of the screw into and out of engagement with the stop abutment, a stop abutment and a runner member each formed with a plurality of complee mentary teeth adapted to engage each other when the runner advances against the abutment, said teeth being defined by radial lines, said teeth having flanks sloping from a point in the axis of the screw and lying at an angle to a tangent to the trace of the screw greater than the helix angle of the screw, the depth and circular arc of the engaging flanks of the teeth being such with respect to the lead of the screw as to cause said engaging flanks to mate with each other over substantially their entire area as the runner member advances against the abutment.

JOHN B. DICKSON.

CERTIFICATE OF CORRECTION. Patent No. 2,5L 9,1LT6. May 16,- 19%.

JOHN B. DIGKSON.

It is hereby certified that error appears in the printed specification of the above numbered patent requiring correction as follows: Page L sec- 0nd column, line 26, claim 5, strike out the words "trace of the" and insert the same before "screw in line 25, same claim} and that the said Letters Patent should be read with this correction therein that the same may conform to the record of the case in the Patent Office.

Signed and sealed this 15th m bf August, A. D. 19%.

Leslie Frazer (S l) Acting; Commissioner of Patents.

Images