US1861155A - Screw pump - Google Patents

Description

May 31, 1932. v DOUGLAS 1,861,155

' SCREW PUMP Filed June 7, 1929 6 Sheets-Sheet 1 20 I 4 I {z 3 i8 I I9 I \KWKA gm INVENTOR EDWIN fi. DOUGLAS M y 1932- 1E; R. DOUGLAS SCREW PUMP Filed June 7; 1929 6 Sheets-Sheet 2 INVENTOR EDWIN RDQUGLAS y 1932! E. R. DOUGLAS 1,861,155

SCREW PUMP Filed June 7, 1929 6 Sheets-Sheet 4 l/Vl/ENTOR EDWIN R. DOUGLAS ATTORNEY May 31, 1932. E. R. DOUGLAS S CREW PUMP 6 Sheefs-Sheet 5 Filed June -7, 1929 L I I I I I I I I l I I I I I I I I l l I l I I I l l I I I l I I l I I I l I l ll INVENTOR fow /v R. UOL/GL A6 ATTORNEY May 31, 1932;

E. R. DOUGLAS SCREW PUMP Filed June 7, 1929 6 Sheets-Sheet 6 lNl/ENTOR ED WIN R. DOUGLAS ATTORNEY Patented a... e1, 1932 UNITED STATES- PATENT OFFICE EDWIN BUST DOUGLAS, OF BALTIMORE, MARYLAND, ASSIGNOR OI ONE-HALF TO HARRY BARKER, F -MONTCLAIB, NEW JERSEY SGBEWIUMP Application filed June 7, 1829. Serial No. 889,087.

This invention relates to rotary pumps, metering devices, or hydraulic motors of that type in which a smaller externally threaded worm rotates and revolves eccentrically inside a larger internally threaded casing, the threads of the two intermeshing and the space between them being filled by a cylindrical rotor fitting the inside of the casing and having an eccentric cylindrical bore in which the worm fits and rotates. If the threads are formed and the parts fitted so as substantially to prevent leakage between them, and the rotor with its included worm be rotated, the device will act as a pump or metering device to pass fluid along the spaces between the threads of worm and easing from one end to the other; or if fluid under pressure be supplied at one end, it will act as ahydraulic motor or metering device, causing the worm and rotor to rotate inside the casing as the fluid passes through.

Such devices have been known for some time, but have not been practical, as the best forms or contours heretofore known for the intermeshing threads have been such as would Wear rapidly and lead to excessive leakage, and also, by reason of their undercut form, have been diflicult to produce with the required accuracy. The characteristic feature of this invention resides in certain forms or contours for the inter-meshing threads which need not be undercut, are easy to produce with the required accuracy, and are not subject to rapid wear or leakage.

In the accompanying drawings, Fig. 1 shows longitudinal and transverse sections through a pump according to the present invention on planes AA and CC respectively,

and Fig. 2 is a diagrammatic transverse sec--- tion on either plane CC or DD of Fig. 1, showing the outlines to a larger scale' Figs. 3 and 4 represent transverse sections of threads suitable for pumps according to the present invention, which have contours based on what is known as the cycloidal system,

and Figs. 5 and 6 represent similar threads having contours based on what is known as the circular or pin-tooth system. Fig. 7 represents a series of four cross sections, by planes parallel to the axis, of one thread de- 5 rived by twisting to an unusually great helical angle the tooth-form of Fig. 4.

In Fig. 1 the numerals 1, 3, 5, represent the spaces between threads of the screw formed in the stationary casing 7. Numerals 2, 4, 6, 55 represent the spaces between threads of the smaller externally threaded worm 8 which rolls around inside the casing 7. The space between 7 and 8 is filled by the cylindrical rotor 9 fitting the inside of the casing 7 and 59 having the cylindrical bore 10 in which worm 8 fits and rotates.

The ends of the casing 7 are closed by the heads 16, 17, the former of which has a packing gland 18 to prevent leakage of li uid around the drive-shaft 19. This drive-s aft passes through the head 16 and is connected to the end of the rotor 9 in any suitable manner so that shaft and rotor will turn together.

In the drawings (Fig. 1) this connection is 7 indicated as belng a coupling 20 of the wellknown Oldham type, operating between the flange 21, which is integral with shaft 19, and the end of rotor 9, so that slight difl'erences in alignment between shaft and rotor will not cause any binding. The outer end of shaft 19 may besupported and driven by any means desired. 1

The rotor 9 is held, endwise, by head 16 through flange 21 and coupling 20, on one end, and directly by head 17 on the other end. The worm 8 is held in position endwise by two flanges 22 and 23, formed integrally with it, which roll against the insides of heads 16 and 17, in a manner similar tothe flange of a car wheel. Thus all end-pressure on rotor and worm is transmitted to one of the heads, 16 or 17.

The space 3 in the casing (either empty or filled with liquid) extends helically around .0

' inside the casing from position 11, where it certain erly together without interference.

a thread of the Worm, to position 15, where it is sto ped off by another thread of the worm. lkewise space 4 in the worm extends helically around from 'tion 12 to position 14, at both of which it is stopped ofi by threads of the casing. If these stop ages at 11 to 15 are tight, so that no liqui leaks ast them, the pump w ll deliver the full vo umetric content of its spaces at each revolution. If they are not made tight or do not remain so, then the pump will be correspondingly ineflicient.

The forms of the threads are for simplicity shown in Fig. 1 as rectangular, but they canis stopped 01! by .not actually be as simple as this, for interferences between them would result at every section between OE and OF, Fig. 2, except on the central section OB. To prevent this, clearances between them must be allowed at oints in certain amounts, and to allow this while still retaining tightness, re-

uires thread contours of certain special apes, now to be described:

For the worm and internal screw to roll to ther, they must bear to each other the re ation of single (or multiple) to multiplethreaded screws of the same helical angle. Practicall the best relation is that of a s ngle-thread ed worm to a double. threaded internal screw, as illustrated in Figs. 1 and 2. For such a pair of screws there may be found two tangent circles GFJ E and OLJM, whose diameters bear the ratio of 2: 1, the smaller circle passing through the center, O, of the larger. As the worm rolls around inside the internal screw, the circle OLJM moves in rfect rolling contact with circle GFJE. These may, therefore, be considered as the pitch circ es of the worm and screw. The

threads may lie entirely inside the pitch circle, or entirely outside, or partly inside and partly outside as shown. Then the nteraction of the threads of the worm and internal screw is the same as that of the teeth of a pair of internal twisted gears of unusually great helical angle. Any forms of gear tooth contours which would work properly together under these conditions would give,

by twlsting to the desired helical angle, screw-thread contours that would work propthem would possess the property of tightness to liquids required for such a screw pump as shown here, but others would not.

There are'several systems of contours used for gear teeth,such as the radial-flank system, the involute system, the cycloidal system, and the circular or so-called pin-tooth system. Up to this time the only one of these applied to screw-pumps of this type has been the radial-flank system. In this system the threads have been entirely inside the pitch circles; the teeth, or threads, of the internal screw then have had straight, radial flanks;

Some of and the threads of the worm are formed to work properly with them. This requires that the threads of the worm be grooved out or undercut.

For this type of tooth, the outside diameter of the internal screw thread must be just twice that of the worm. This combination in such a pump forms the subject of a U. S. Patent, #1,490,876, issued to J. K Werner, April 15th, 1924.

Theoretically, threads of this type would be liquid-tight, but those contacts which do the sealing are sharp corners, like knifeedges, and rub against fiat surfaces. The sharp corners of contacts like these will very soon wear off round and allow leakage. Moreover, the undercut curved form of contour, which must be very accurately fitted if tightness is to be secured, is not only diflicult to make but especially difficult to make accurately. Hence, threads like those based on the radial-flank internal gear tooth are not practical for service. These, however, constitute the state of the art covering pumps of this type prior to the invention herein claimed.

As before mentioned. Figs. 3 and 4 show two tooth-forms of different proportions based on the cycloidal system. and Figs. 5 and 6 show two tooth-forms based on the circular system, and Fig. 7 shows a series of four cross sections of one worm thread and screw space derived by twisting from the tooth-form of Fig. 4, these sections being made by planes OB, ON 0N and OE of Fig. 4, respectively. Similar series of cross sections could be made for all the tooth-forms of Figs. 3 to 6. but the additional ones would be so similar to those of Fig. 7 in all but unimportant details and proportions that they are not shown.

All the forms shown in Figs. 3 to 7 have a common property: The contacts between the surfaces of the worm and internal screw are tangent surfaces presenting good wearing contacts, like well-designed gear teeth. instead of sharp-edged contacts as in the radialflank system. They differ also from that form in that the teeth, or threads, lie partly outside of the pitch circles, so that the ratio of the outside diameters is no longer 2: 1. In Fig. 3 it is7z4, or 1.75: 1; in Figs. 4 and 6 it is 29:17. or 1.706:1; in Fig. 5 it is 9:5, or 1.8: 1; and other proportions might be used.

In all these diagrams Figs. 3 to 6. O is the center of the internal screw, which has its pitch circle J GE described about 0; and K is the center of the worm, which has its pitch circle JMO described about K. The path of contact between the teeth of the two is ZJCDH.

In Figs. 3 and 4, on the cycloidal system. the tooth-contours are generated by the internal generating circle IZ'JZ", having its center at V, and the external generating circle J Y having its center at W. If (as shown these two circles, VJ plus JW, equals the radius J'K of the pitch circle of the worm, then, as is well known to those familiar with the theory of gears, there is also an intermediate describing circle J CDH, having its cJeVIger at X, such that XK equals KV equals The internal circle IZJZ" generates the flank J Q of the worm contour and the face JP of the internal screw contour, which maintain tangential contact with each other to the right of J along contact circle IJZ", from oint J to a oint Z, not shown, where circle J Z", exten ed, intersects circle HZ'P, extended. At these points, J and Z, this particular tangential contact terminates. In like manner the external circle J Y generates the flank JT of the internal screw and the face JF of the worm, which maintain tangential contact with each other to the left of J along contact circle J Y from J to Y, at which points this contact terminates. Furthermore, the face JP and face J F are again enerated by intermediate circle J CDH and t iese faces maintain tangential contact with each other to the left of J along contact circle J CDH from J to H, at which points this contact terminates. In addition to these three tangential contacts there are the followin nontangential contacts. The corner of t e internal screw tooth, P, maintains non-tangential contact, as at S in Figs. 8 to 6, with the root of the worm tooth, which is shaped to conform, in the curve QR which forms a fillet at the base of the worm tooth. This non-tangential contact extends along circle HZ'P from oint P on the center line toward the rig t, to point Z, not shown, but mentioned above. This latter point, as already seen, is also the terminus of the tangential contact JZ"Z. Also the corner of the worm tooth H maintainsexactly similar non-tangential contact with the root of the internal screw thread, the latter being shaped to conform in the curve TT which forms a fillet at the base of the internal screw tooth. This non-tangential contact extends along circle FUYH from point U, on the center line, to point Y, which latter point is also a terminus of the tangential contact J Y.

If an internal gear and pinion made on this system be rotated together in clockwise direction about their centers 0 and K, their teeth will first come in contact at the point Z (not shown) to the ri ht of Z on circle- IZ J Z" and will remain in contact until they reach point H. There will be continuous tangential contact along curves ZZ" J, JCDH.

and JY. From J to C and Y there will be double tangential contact, as shown atC and Y. There will be continuous non-tangential contacts from Z to P and from U to Y.

Every tooth must have two sides, only one of which is indicated here. For the other side, similar contacts would 'result, exee t that they would be reversed, right and le t, about the center line XW. If such an internal gear and pinion be twisted uniformly to ive an internal screw and worm similar to Fig. 1, the two sides of any mating gear tooth and space will twist into the two sides of a mating screw thread and space, and their successive transverse sections will give continuous contact between the helical surfaces of the screw worm, all along the curves ZJ, JCDH. J Y, ZP and UY on one side, and also along the corresponding (left handed) curves for the other side. If, for exam le, the twisting be such as to produce right handed screw threads (in which for clockwise rotation successive sections move away from: the observer), then the contours shown in Fi s. 3 and 4 would produce the nearer side 0 a mating worm thread and internal screw space; while the corresponding left handed contours would reduce the farther side.

This has been in icated in Fig. 2, where the broken-line curves JH, J Y and JZ indicate the lines of tangential contact and lines ZP and YU indicate lines of non-tangential contact for the nearer side of a worm thread and internal gear 5 ace, and the dotted-line curves JH', J Y, indicate those for the farther side.

The continuous and tangential nature of these contacts is further indicated in Fig. 7. As already stated, this shows a series of four cross sections of one worm thread and screw space derived, by twisting from the toothform of Fig. 4:, these sections bein made re spectively by planes OB, ON,, 0 and OE. All these planes pass through the center line, 0, of the Internal screw, andhence the shape of these cross sections, so far as the internal screw is concerned, will be the same in all of them. But only plane OB passes through the center line K of the worm. The other planes cut this worm at successively greater distances from its axis. Hence the sections of worm thread made by these planes will all be unlike.

In Fig. 7, the cross section of internal screw space, alike for all four planes, is P J B,, B1 J1,- P1- The four cross sections of worm threads made by these four planes are as follows:

The right hand side of Fig. 7 corresponds to the nearer side of a right-handed worm thread and screw space twisted from the tooth contours of Fig. 4, and the left hand side'of Fig. 7 corresponds to the farther side of such a thread.

The apparent skewing of the worm thread sections in Fig. 7 is due to the section planes cutting that thread at successively greater obliquities. There is, of course, no change in the actual sha of the thread.

Asthe sectioning lane swin to the left in Fig. 4, passing fi'om OB t rough ON 0N and all intermediate positions successively to OE, the worm thread is shown in Fig. 7 to lift out of the internal screw space. As it does so, the sides of the worm thread are seen to maintain contact with the sides of the internal screw space in a continuous and tangential manner. On the left hand side they do this for a time only, after which conars, but on the righttact on this sidedisap hand side contact is ully maintained until the worm thread is all the way out. Starting with section by plane OB, there is contact on the left at J, and on the right at J (corof Figs. 2 and 4), where the worm thread is fully out of the internal screw.

The above contacts were all tangential. There is also a continuous non-tangential contact at the left between corner P, and fillet curve R.Y., as at point S of the worm tooth. This non-tangential contact S lasts until tangential contact Y, has moved up to corner P,, when both of them disappear. There is also a continuous non-tangential contact at the right between worm tooth corner and the fillet curve from B, to Y. This lasts while thefivorm tooth corner is moving up from U to The character and extent of these tangential and non-tangential contacts between the surfaces of worm and internal screw, asshown in Figs. 2, 4, and 7, has now been made plain. It remains to show how they close of the successive spaces of the pump from each other.

In Fig. 1. section C C cuts through spaces 3, 4 and 13. At the latter, the worm thread meshes into the internal screw space and separates internal screw space 1 (coming down helically around the front) from space 5 (going up helically around the back) section C C cuts these parts of spaces 1 and 5 also.

If the diagrammatic cross section in Fig. 2 be ascribed to section C C of Fig. 1, then it will show these spaces. They are marked with a sub-script c to indicate that they belong to section C C of Fig. 1. They are: space 3.. at the top of the internal screw, space 4, at top of worm, spaces 1 and 5 at left and ri ht of center line, and 13. at the bottom. Hire the worm thread fits down into the internal screw space and the two are in close, tangential contact, on the nearer side along line HJ Y, then across the bottom on line U,

farther side along the line YJH. These would effectually shut of! space 1 on the left from s ace 5. on the right, were it not that from 1 to U on the nearer side and from U to Y on the farther side of the contact is not tangential and there is a possibility of sli ht leakage.

' On the nearer side, li uid from space 1., could pass under the edge of Fig. 2 (which is line U, U, of Fig. 7) up around the nontangential contact edge YU (YU in Fig. 7)

around through the thin space between the line YU and contact line YJ to line JU of Fig. 2 (which is the narrow triangular space J B U in Fig. 7), and so on through to the right into space 5.. This will be called leakage a.

On the farther side there would be an exactly similar and equal leakage from s ace 1. throu h the narrow triangular s ace 2 13 U, of 1g. 7, down around the e ge U, Y under Y and so on into space 5.. This will be called leakage b.

Again. in Fig. 1, section D D cuts through space 12. Here the internal screw thread meshes up into the worm space and separates worm space 2 (coming down around the front) from space 4 (going up around the back) Section D D cuts these parts of spaces 2 and 4 also.

If the diagrammatic cross section in Fig. 2 be now ascribed to section D D of Fig. 1, then it will show these spaces. To differentiate them from the ones previously considered they are marked with a subscript (1. They are: space 2., at the left, space 4 at the right, and 12., at the bottom. Here the internal screw thread fits up into the worm space and the two are in close tangential contact, on the nearer side along line JZ, on the farther side along line J Z. and across the top on line P, from P to P, of Fig. 7.

There are here two possibilities of slight leakage from space 2., to space 4,. On the nearer side of space 12, liquid from space 2., could pass down around the non-tangential contact edge Z P (P in Fig. 7), around through the thin space between the line Z P and contact line Z J to line PJ, Fig. 2, (which is the narrow space P Q. J 2 in Fig. 7), and so on into space 4 .This will be called leakage c.

On the farther side of space 12 there would be an exactly similar leakage from space 2., through the narrow space similar to P, (.2 J, of Fig. 7, up around the edge PZ. and so on into space 4.. This will be called leak age d.

Fig. 7 represents spaces 12, 13, and part from U1 to U, of Fig. 7, and then on the of 14, of Fig. 1. At space 12, the internal screw thread separates worm space 2, conr ing down helically from the left, around the front, and worm space 4, going up helically around the back, to the right. At space 13,

the worm thread separates internal screw spaces 1 and 5 as already described. At space 14, the internal screw thread separates worm spaces 4 and 6 in a manner similar to that at space 12. Then it is seen that, at the left, spaces 2 and 1 are in full and open communication through the opening to the right of edge P This in Fig. 2 18 all along the line Z H at the farther side.

Similarly, of course, if spaces 1 and 2 are in free communication with each other, so will be spaces 3 and 4, and .likewise 5 and 6, of Fig. 1.

At the right, in Fig. 7 it is seen that space 4 is always separated from space 1 by the close tangential contact, as at J Y D, and P etc., which is the same as contact line JH at nearer side in Fig. 2. There is no leakage path here.

To recapitulate: Spaces 1 and 2 are in free communication and form one space; similarly with spaces 3 and 4, and with spaces 5 and 6. There is no communication or leakage path from space 1 to space 4, nor (similarly) from space 3 to space 6. Space 1 is shut off from space 5, on the nearer side, by the tangential contacts along lines HJ and J Y and the non-tangential contact along line YU; by the tangential contact across the bottom at U and on the farther side by nontangential contact UY', and tangential contacts YJ and JH. There is possibility of a slight leakage (a) past non-tangential contact YU and (b) past non-tangential con tact UY'.

Space 2 is shut off from space 4 and, in similar manner, space 4 from space 6 on the nearer side by tangential contacts on lines HJ and JZ and non-tangential contact ZP; across the top at P and on the farther side on non-tangential contact PZ' and tangential contacts Z'J and JH. There is possibility of a slight leakage (0) past non-tangential contact ZP and (d) past non-tangential contact PZ.

These four possibilities of leakage, a, b, c, d, from space 1 to space 5 and from space 2 to space 4, due to the non-tangential nature of contacts on edges YU, UY, Z'P and PZ are, however, very small. Even if the corners at roots of worm and screw teeth were sharp reentrant corners entirely unfilleted, the openings of these edges would be very small. Taking as unit area the total combined area of space 12 plus space 13 (which is depth of thread pitch of single thread and space) such measurements and computations from Fig. 4 yield the result that this combined leakage cross-section past corners if unfilleted would equal only 0.0028 of the combined working cross-section of the threads, or less than of 1%. 7

With lilleted corners giving continuous contacts, even though sharp-edged ones, this leakage is, theoretically, completely stopped.

In practice, due to slight inaccuracies and effects of wear there might be slight possibilities of leakages past these corners. However,

these contacts have no mechanical work to do. That is all done by the tangential contacts of the tooth faces. The wear on these non-tangential contacts will, therefore, be negligible and they will retain their original tightness for a long time. The possibility or leakages past them is further greatly reduccd by the very thin spaces through which such leakages must also pass such as around from YU to J U and from ZP to PJ, in Fig. 4. Hence the total leakage past these nontangential contacts will be but a very small fraction of the of 1% mentioned above and is quite insignificant; This is in strong contrast to the eIIects of wear in non-tangential edges heretofore used, as shown in U. S. Patent 1,490,876 previously referred to.

The above discussion has been concerned with the contours of Figs. 4 and 7, but would have been the same for Fig. 3, which is also on the cycloidal system, except that the ratio of leakage to total area might have dilfered by a small fraction of a percent.

The dili'erence between the contours of Figs. 3 and 4 lies in the proportions of their generating circles. in Fig. 5 these are chosen to give equal height and depth of tooth above and below the pitch circles. This results in giving a slightly undercut pinion tooth, as shown at A, Fig. 3. 111 Fig. 4 the generating circles have been chosen to give'no undercutting at A, and this causes the tooth-depth to be greater outside than inside the pitch circles, which is in no way detrimental.

The contours of Figs. 5 and 6 are generated in a different way, but lead to practically the same results. ln these, the working contours of the pinion teeth are circles, such as QJ U, struck from centers lying on the pinion pitch circle. One of these centers is shown at W, for the contour circle passing through H. Then the contour of the internal gear teeth outside its pitch circle will also be circles of the same radius as those of the pinion, but struck from points of the pitch circle of the intcrnal'gear. Inside its pitch circle, its contours will be straight lines such as JP, tangent to the outside contour circles at points a little outside the pitch circle.

- Then the path of contact will be a curve (not a circle) ZJCDH, which is generated by a point F on a line JXF passing through the point of tangency J of the pitch circles, intersecting the pinion pitch circle in X, and with the distance XF kept constant and equal to the radius VVH of the contour circle as X moves around the pitch circle J XM.

When the center of the contour circle is at J, the pinion contour circle and the internal ear contour circle will coincide at CY, an when the teeth are twisted into the form of screws this curve CY performs the same functions as did the contact line J Y in Figs. 3 and 4: It gives a line of tangent contact between the nearer surfaces of worm and internal screw threads and, in conjunction with contact line J CDH, shuts off space 1 (at the left) from space 5' (at the right) down to oint Y. Vith this minor (litterence, the iscussion for Fig. 4 will apply also to Figs. 5 and 6. The difference, if any, in the leakage percentages, would be of a minor order.

The difference between the contours of Figs. 5 and 6 is similar to that between Figs. 3 and 4. InFig. 5 the radius of contour circle, WH, has been so chosen as to give greater tooth-depth inside than outside the pitch circles, but this leads to slightly undercut pinion teeth, as at A, Fig. 5. In Fig. 6 the radius \VH has been so chosen as to give no undercutting at A, but the tooth-depth outside the pitch circles is now greater than that inside, as it was in Fig. 4.

It is evident that other roportions could be used for the pitch circ es than those in Figs. 3 and 4, or other radii for the contour circles than those in Figs. 5 and 6, without changing the nature of the result greatly. These might even be so chosen that the teeth would lie either entirely inside or entirely outside the pitch circles, provided the lines of continuous tangential contacts between wearin surfaces were afforded. It is also probab e that from the infinite number of tooth contour systems possible, other systems than the cycloidal or the circular types here illustrated could be found that would give substantially the same results. This invention does not reside in a particular kind of tooth or thread contour, but in the application, to this type of screw pump, of any of the forms of tooth contours that will give broad tangential contacts between the wearing surfaces of the threads of the worm and internal screw and maintain substantial liquid-tightness at the barriers that separate their successive spaces.

What I claim is:

1. A screw pump having a threaded screw member eccentrically positioned within and meshing with an internal thread of an enclosing member having an internal diameter relatively greater than the outer diameter of the screw and in which the space between the screw and the internal thread is filled by a rotor member, characterized in this; that the contours of the intermeshing threads are pairs of curves which are the intersection of a longitudinal axial plane with pairs of intermeshing helical surfaces generated by twisting about their respective axes an internal gear and its intermeshing pinion whose tooth surfaces are of that particular class that make continuous tangential surface contact with each other from their first to their final contacts.

2. A pump according to claim 1 characterized in this: That the thread contours are of a form which would result from twisting an internal gear and its intermeshing pinion whose teeth are formed according to the cycloidal system.

3. A pump according to claim 1 characteu ized in this: That the thread contours are of a form which would result from twisting an internal gear and its pinion whose teeth are formed according to the circular system.

4. A pump according to claim 1 characterized in this: That the outside diameters of the threads of the internal screw and the worm are greater than the diameters of their pitch circles.

5. A pump according to claim 1 characterized by this: That the contours of the intermesliing threads are of a form which would result from twisting an internal ear and its pinion whose pitch circles have diameters in ratio of any two whole numbers.

6. A pump according to claim 1 characterized by this: That the contours of the intermeshing threads are of a form which would result from twisting an internal gear and its intermeshing pinion whose pitch circles have diameters in ratio of any two whole numbers and whose tooth contour outside diameters have a different ratio than their pitch circles.

7. A screw pump having a threaded screw member eccentrically positioned within and meshing with an internal thread, of an enclosing member having an internal diameter relatively greater than the outer diameter of the screw and in which the space between the screw and the internal thread is filled by a rotor member, characterized in this: That the contours of the intermeshing threads are of a form which would result from twisting an internal gear and its intermeshing pinion whose pitch circles have diameters in ratio of any two whole numbers and whose tooth contour outside diameters have a different ratio than their pitch circles.

8. In a screw pump, metering device, or motor, the combination of an internally threaded member, an externally threaded worm eccentrically arran ed within the said internally threaded mem er, the threads of the two having the same helical angle and intermeshing with each other, a rotor having a clindrical outer surface fitting the inside 0 the internally threaded member and having an eccentric cylindrical bore fitting the outside of the worm, the contours of the said intermeshing threads being of a form which would result from twisting an internal gear and its pinion whose pitch circles have diameters in ratio of 2:1, their teeth being formed on any tooth contour system that provides continuous tangential surface contacts between their Wearing surfaces from their initial to their finalcontact.

9. A pump according to claim 1 further characterized in this that the re-entrant corners of the mating spaces of said gear and pinion are filleted in such a manner that the projecting corners of the gear and pinion teeth make continuous non-tangential contact therewith from their first to final contact. I

EDWIN RUST DOUGLAS.

Images