Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.


  1. Advanced Patent Search
Publication numberUS1793660 A
Publication typeGrant
Publication dateFeb 24, 1931
Filing dateMar 3, 1927
Priority dateAug 17, 1923
Publication numberUS 1793660 A, US 1793660A, US-A-1793660, US1793660 A, US1793660A
InventorsWood William H
Original AssigneeWood William H
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Process of making punches for reflectors
US 1793660 A
Abstract  available in
Previous page
Next page
Claims  available in
Description  (OCR text may contain errors)

Feb. 24, 1931 original FiledfAug. 17,.-1923 2 sheets-sheet 2 wniam www@ Accorneys and improved p jects of the invention relate to the producobjects an PatentedlFeb. 2 4, 1931 UNITED vs'rnras PATENT o1=1=1c1sA A WILLIAM H.

woon, or soUTH EUCLInoHIo ,rnocnssv or MAKING PUNoiins For. BErLEcroRs Original application filed August 17, 1923 SerialiNo. 657,996. Patent No. 1,868,706, dated May` 8, 1928.`

Reissued December 4, 1928, No. 17,155. Divided and this application led March 3, 1927.- Serial No. 172,312. This invention relates to vehicle headlights ,tion of myl improved reflector punch; Fig. 5

and has forits object the provision of a new unch and die whereby a onepiece sheet metal reflector' can be simply and easily made which shall produce a legally ac.-- ceptable and practically satisfactory distribution of light without the use Iof lenses,` prisms, shades, dimmers or other expensive, fragile, or light killing devices. vOther ob- `represents the second view of the punch sh illustrates th step; Fig. 6 is a face the conclusion of face view of the ustr'ates the same first cut taken off face view of the punch shown in Fig. 9; Fig. llfshows the` same punch after having had thenext cut', taken Fig. 12 is a front view of the punch shown in Fig. 11; also illustrating the nature of the next succeeding step; Fig. 13 is a side elevation of the completed punch; Fig. 14 is avertical central section of a reflector made a front elevation of the Fig. 14; Fig. 16 is a face ring m e same punch at the next step; Fig. 8 is a punch in Fig. 7; Fig. 9 ill punch after having had the its upper part;vFig. 10 1s a tion of such a punch and di gle piece of metal; the pro of such construction that it can readily be dressed or re-shaped the provision of'a device of this character which can be very quickly and cheaply produced; the provision of an improved headlight reflector, while further d advantages of the invention will become apparent as the description proceeds This application is a division of my application filed August 17, 1923, Serial No. 657 ,996, issued May 8, 1928, Patent N o. 1,668,706, reissued December 4, 1928, Reissue No. 17,155.A In my former application filed Jan. 24, 1923, Serial No. 614,694 I describe a certain mathematical surface for use in the construction of headlight reflectors for producing'a` light which is not only legally satisfactory, but also to aconsiderable extent independent of minor variations in focal distance. The present invention contemplates the use of these curves, or other curves 'similarly found, but refers specifically to the manner of inipressing these curves upon the metal block whereby the tool itself is-produced. In the drawings accompanying and forming a part of this application wherein I have shown in detail onemode of producing my improved tool; Fig. 1 is a diagrammatic view showing the inode of generating my improved curve as employed for 'the top portion of my reflector; F ig. 2 shows the mode of generating my improved curve as employed for the lower part of my reflector; Fig. 3 is a diagrammatic view showing the preferred mode of assembling these curves togetherA and disposing them with reference to the horizontal, this diagram conforming substantially to the central vertical section of my preferred reflector; Fig. 4 illustrates the first step in the preparae each from a sinvision of a punch reflector shown in view of the die or the punch co-operctionalview of the ember with which ates; and Fig. 17 is a se ring shown in Fig. 16.

y improved curve consists in effect of a plurality ofjparabolas merging one into the other, each portionl corresponding to a differ'- ent focal point. The shape of the curve forl the top half of the refiect ferent from that of lhe l veloped as follows: I d Fig. 1 to represent the above it I locate the point S say one-fourth inch therefrom, this distance corresponding to the intended elevation of the light source above the reflector axis. vIt will be understood that' this dis less or greater lthan the amount I have suggested, and will also depend upon thefocal depth of the curve. If made too great, difficulty will be experienced in merging together thedifferent surfaces inasmuch as they will be severely warped;'if made too small, the desired light control willnot be obtained. The present drawings are exaggerated in all respects in order to show the theory of operation within the small scale permitted. Through the point S I draw a vertical line two inches long ending at. the axis at 2. This line correspond ovv'er` part and is' deraw the line 1--1 of at which its inclination to the axis 1 1 is thereon; Fig. 15 is l own in Fig. 5; Fig. 7 I

or is materially difhorizontal axis and v placement may be either s to the parameter of a parabi ola and its upper end 3 is a point of the curve at the same distance substantially 45. Also at a point in the rear would be followed if a parabola were to be drawn about the point 2, with a focal distance of one inch. Of course I do not limit myself to this size.

However, if the point Sbe considered as a light source, lthen itsvirtualposition along the axis I will depend upon the angularity with which the rays strike that axis. For example I now draw an inclined line through the point S so as to intersect both, the axis 1 1 and the probable curve, say the line e intersecting the axis at 4. I then draw a perpendicular that 4 lies .in front of 2 and on the line e I locate the point 5 which is equidistant from the line E E and theintersection 4. In other words thepoint 5 constitutes a point on a parabola whose true focus G Gr as 4far in front'of the point 2. On this line I then lines to plot an accurate curve between the parameter is E E.

another inclined line g point S intersecting the axis at draw a corresponding directrix behind D D as the point 6 lies is 4 and directrix I then draw through the 6, and also find the point 7 which is equidistant from the point 6 and from the line G G, the point 7 being upon a parabola whose focus is 6 and whose directrix is G G. I repeat the same performance for a sufficient number of other locate other points 8, 9, etc. and thus and the vertex.

I then draw an inclined line 7' through the point S intersecting the axi-s at 12 between the parameter and the vertex and to accompany it I draw a. perpendicular .I J as far in front of the lin'e D D as 12 isv behind 2, after whichI lindthefpoint' 13 on this line which is equidistant from the point 12 and the line .I J, this'point being` located on a parabola whose focus is 12 and whose directrix is J J. This is repeated for the lines 7c and K K etc. and giving points14, 15 and as many other times as desired after which a curve Y is drawn connecting the points 3, 5,

To plot the line below the axis I draw a second 'straight line 20 20 as in Fig. 2 and place at one-fourth inch above it (or whatever distance I previously employed) a point S from which I drop a perpendicular p intersecting the-axis at 21; and on this line at about two and one-fourth inches (more or' less, but preferably 'a littlemore than the distance 2 3 for a reason to be explained hereafter) below the line I locate the point 22 which is a point on the curve. Also I drop the perpendicular? P so located that the peating this with other lines r,

E E the same distance behindthe line D D point 22 shall be equidistant from 21 and from P P. i l v I then draw an inclined line g through the point S intersecting the axis 2O at 23 behind the point 21 and I also draw a second perpendicular Q Q the same distance in front of P P as 23 is. behind 21. On the line g I then locate the point 24 equidistant from the point 23 and from the line Q Q. VAfter ret, etc., I find the points 2G, 28, etc., sufiicient in number to define an accurate curve. I draw other lines as u, and o intersecting the axis in front of the point 21 and for each I draw a corresponding perpendicular as U U, V V an equal' dist-ance behind the line P P -thus wherein p is the abscissa and c the ordinate of the point S or S', but it can be demonstrated that throughout avconsiderable arc i the same .is practically indistinguishable from the portion of a parabola lying between its vertex and parameter, the axis of said parabola passing through thepoint S or S and inchned forwardly and downwardly so as to make an oblique angle with the con-v struction axes 1 1 and '20 20 respectively. Such parabolic axis is-indicated at 100, in Fig. 1 and 101 in Fig. 2.

According to my referred species the reflector' to be made roin these curves has a central vertical section made by assembling these curves as shown in Fig. 3, the line 1 1 being super-posed upon the line 21,) 20 with the point S located slightly behind the point S.4 If the distance 21, 22 [was properly chosen relative to the distance 2 3, the vertices of the two curves will match properly with the desired longitudinal separation of these points which is preferably rather small-v say fg inch (it will be understood that the present patent drawings are entirely outof proportion in order to produce suliicient separation ofthe different points and lines for ready reference). If a light source A be located at any point'between S an S', all rays falling on Y will be reflected forwardly and downwardly as shown at B. Indeedthis downward deflection is so great as to render it desirable to tip the reiiector upwardly as regards the horizontal line H H. The dotted line F in Fig. 3 indicates a true parabola drawn about the axis 20 for purposes of comparison. ,I will now describe the mode whereby I apply my improved curves to the metal tool whereby this reflector is made. I irst take a suitable. block of metal shown by the dotted no l ' turn a cylindrical shaped metal conform to the surface of revolution of ,the

portion of the block is formed by a of revolution of the portion Z of the curve' outline in Fig. 4 and on the one side I boss 41 while the opposite already described, .theaxis 2 0-20 being coaxial with the boss-4l and with the axis of revolution'. This produces a mushroombodily laterally to a new axis ofrevolution parallel to but spaced from the axis and the block is again machinedto same curve about the axis` 42 with a radius equal to slightly less than the sum of the displacement 42 plus the previous radius. While in this position I cut into the block at one side'throughout-the are indicated approximately at 43-43 in Fig. 6, there y slightly flattening the .block on one side and taking off a comparatively thin layer indi cated at the bottom of Figs. 5 and 6. This accomplished I-shift the block still further in the same direction and again turn it about the axis 45, parallel tothefirstaxis but displaced yet further in the same direction,

n and cause the same to conform throughout `a yet narrower area to the surface of revo- A properly chosen,

lber of settings is unnecessary.

lution of the same curve `about a radius similarlyincreased. ,In this way I remove a yet narrower region represented by the arc l46-46 in Fig. 8. A still greater numcan be made if desired, but Indeed I have had lexcellent results with only a single offset cutting.

I now oliset the block in the opposite direction to an axis 48-48 and machine it to conform to the, surface of revolution of the curve Y about this new axis, the radius ofrotation being. increased by the distance be-v tween 20 and 48. If this displacement is' tom portion of the block swinging inside this new surface -of revolution. In the embodi- 'ment here illustrated the curve Yv is much smaller than the curve Z owing to the comparatively great vertical displacement of thepoints S, S', and for this reason the vcut 'extends more than a halfcircumference, from 49-49. In actual practice with the meas- 1 .urements here indicated the cut will not ex- -tend over even a semi-circumference. Either condition is equally satisfactory since the side wings hereafter described will fill the gap if such exists.l Thereupon the block 1s again shifted to an axis 50 yet further displaced and is creased radius,-

only a comparatively narrow strip being c ut along the extreme ltop of the punch as indicated by the arc 51-'51 in Fig. 12.- y VThe block is now shifted to a new and very eccentric axis 52, located somewhat below the axis 20 and very materially to one sideand surface block which is now shifted l 'the punch along this line;

'shall -conform closely therewith,

the block will be surfaced over only a part of its circumference, the botagain machined to the same 1t surface of revolution about a simllarly m-A- the lines 20-20 and la side wing turned throughout the e are 53-58 extending from thc largest portionof the die to a pointsubstantiallyopposite the parameter. This wing will appearlslightly liattened and somewhatl wedge-shaped as indicated at X in Fig. 1,3; after which the block is again shiftedtothe axis 54 symmetrically All these Avarious sections, if the4 work-has ocated as regards the axis 52 and corresponding to the arc,55.-55 f beencarefu'lly done as herein directed, will merge together with very little unevenness. It will be necessary to smooth down themeeting edges `with a fine iile necessary in this regard is notv to dress the `same too much, merely to -remove roughness and produce a smooth blending. The block. plate`56 shown in Fig. 13 having a shank 57 adapted to it thev 1s now mounted in a base socket of the punch press plunger, the face and back of this base plate being inclined the amount by which the axis of the lamp. is to diverge from horizontal. The punch is removably secured to this plate mit redressing in case of need.

Owing to this inclination the punch will strike the Vdie at. a slightly different angle from that originallyexhibited namely along the line `60. This die, therefore, consists having its inner merely of a steel ring 61 face 62 shaped to conform to the section of Common boiler plate is amply suicient for this ring. Y The absence of anyre-entrant angles or vdepressions inthe surface of the'punch dispenses with the need for a receiving member which in other Words a bottoming die.' .The reflector' produced by this tool` is illustrated in Figs. 14 and 15, and is made'by a single stroke of the 7 tool, the edges of the 'plate being meanwhile gripped against the ring.61 inthe manner so as Lto perbut the only care well known to all those skilled in the art of metal drawing. A portion of this metalnis left as at 65 to form a'stifening rim or flange whose exterior is trimmed to circular shape, while the heel of the reflector is' provided with an apert'ure W for the lamp socket, the same being located suiiicienty above the axis 20-2() to bring the light source the desired distance therefrom.

It will vbe understood that I do not limit myself to anyone relation of distances or ratio between the upper and lowei` curves or v displacement of the foci of said curves or in anywise except as reception of the producing the' original blankI ter is equal to the sum of the cuts which are to' be taken off the top and bottom of .the blank. If this be done at the start, then the inahpunch i will exhibit the exact section shownin Fig.l 3; if it be n ot done the device Willbe narrowed by this amount, although if.

,properly designed the light will 'still be acceptable. Thls is only a Working allowance such -a's any carpenter must provide everyA tnne'he saws a board,.but my invention 1s not restricted to -such niceties although the accuracyof the result is improved thereby. Also I have described a device made by swinging the blank about numerous axes 20, '42, 45,

.'tice since common on the face plate, but I do not limit myself` settings, having shown,

48, 50, 52, 54, all of whichare parallel to each other which is extremely convenient in prac'- it permits the tool to be madeon a lathe merely by shifting the block to'this parallel arrangement. AlsoI do not limit myself to the use of all the curves and as in some instances it may be feasible to dispense with the side wings or other elements; and conversely l may use additional curvesA and additional settings.

Having thus described my invention what Iclaimis' Y 1. The process of making a punch forreectorswhich contains the steps of forming a metal block at least 'in part as a surface of revolution of a segment of a parabola vabout an axis which is displaced fromthe focal point of such parabola, then displacing said `block laterally and forming one side thereof asa partial surface of revolutionof a segment of a second parabola about a second axis which is eccentric to the first axis.

tion which is oblique to the parabolic 2. v-The process of making a punch for reflectors Which contains the steps of forming one portion of a metal blockto conform to a part of a surface defined 'by revolving a segment of a parabola about an axis-of revolution Whichis oblique to the parabolic axis,

4g and forming a second portion of said block to conform to a part of a surface defined by revolving a segment of a second parabola 4about a second axis of revolution which 'is oblique to the geometric axis of said second arabola,` both of said axes of revolution lying in the same plane.

3. The process of making a punch for reflectors which contains the steps of forming one portion of amctal block to conform to a part of a surface defined by revolving a seg-A ment of'a 'parabola about an axis of revoluaxis and'displaced beyond its focal point, and forming a second portion vof said block which is on the opposite side of said rst axis of revolution from the first described portion toconform to a partial surface of revolution defined by rotating a segment of a different parabola about an. axis which is displaced from said first axis.

.axls of the first generator, each of said generators subtending only a portion of the parabolic curve which defines the same.

5. The. process'of makinl a punch for reflectors which comprises orming different portions of a metal block as surfaces of revolution about different axes of generators consisting of plane curves whiclr have focal points and axes, the axes of revolution 'being located. further their o Wn geometric axes and also lbeing obliqueto such geometric axes.

In testimony whereof I hereunto affix my signature.


from said generators. than

U.S. Classification76/107.1, 362/347, 82/18
International ClassificationB21D53/88, B21D53/00
Cooperative ClassificationB21D53/883
European ClassificationB21D53/88B