US2298808A - Television projection system - Google Patents

Description

E. G. BAMBERG 2,298,808 TELEVISION PROJECTION SYSTEM Filed Ap'ril 2e, 1941 /8 r -fa-.5. M/ROR n 7 Oct. 13, 1942.

4 Sheets-Sheet 2 Rlvy 1N Mamma/wiz; i PLANE Edivad G. Ramberg E. G. RAMBERG TELEVISIONJPRQJCTION SYSTEM "Oct 13, 1942.

4 Sheets-Sheet 5 Edward G. Bamberg rllvlll' Si os W if.. ...i

Oct. 13, 1942.

E. G. RAMBERG TELEVISIN PRQJECTION SYSTEM Filed Au 25; 1941 4 Sheets-Sheet 4 I Il 1.9-

4 :inventor Edward @.Hamery PatentedOct. 13, i942 'miran stares fLTl-:NT

of Delaware Application Apriizs,1941,seria1N.390,4 .'z.ciaims...-fCl.17s -7. 5) My inye'ntion relates to systems for projecting television pictures or the like and particularly to projection systems in which the optical system is of the type comprising a sphericalv mirror and a spherical-aberration correcting plate.

As explained in application Serial No. 248,569, led December 30, .1938, in the name of Daniel O. Landis, it has been found that alarge television y"arcanos l rELvIsroN rnomc'rrou SYSTEM Edward G. Ramberg,-Moorestown, N. ..,`asslgnor to Radio Corporation o! America, a corporation picture can be projected with suicient illumination and with good denition by employing a specially designed optical system of the abovedescribed type. The present invention is an improvement on the system disclosed by the afore- When theliirst television projector made in accordance with the Landis invention was completed and tested by projecting a picture from` only andwhen figured in accordance'with-t present invention, v

Figure 3 is' a-diagram which is referred inl'- describing the error introduced by the-thickness, Y

of thetube face, I i

Figure 4 is'adiagram which is referred to with o- Vreference to formulas employed -in tracing a' -paraxial ray, A Y

Figures 5 and 6 are diagrams, not to scale, which are referred to in explaining how a 'paraxial ray and a ray at a large angle to the are traced through the sys;

kwhich arel traced through the system in order,

' to select the preferred values 'of tube face radius vand of focal length for the center of the correcting plate, and

Figures 8, 9 and 10 are curves which are calculated and utilized for selecting the above-mentioned preferred values of radius and focal'length.

the end of a cathode ray tube, it was found that, c,

while the projected image was much brighter than that obtained by any previously employedoptical systems, theV detail -was less than had been expected for the known degree of accuracy of the system. v

I have discovered that the end of the cathode ray tube envelope, that is, the tube iace, introduces an error in addition to the spherical aber-l ration error of themirror which noticeably impairs the deilnition of the projected-picture unless an additional correction is made. In other words, while an image of good deiinition may be projected by fully correcting onlyfor the spherical aberration of the mirror, the quality of the image -will be improved and will 'be of a very high order if the additional correction for the tube face thickness is made.

In accordance with my invention, this error is corrected by shaping or guring the correcting plate of the optical system to correct not only for spherical aberration of the mirror butnalso for the error introduced by the thicknessof'the end of the cathode ray tube.

The invention will be better understood f ro the following description taken in connection with the accompanying drawings in which Figure l is a side view of a television projection system embodying my invention,

Figure 2 is a sectional view of the correcting plate of the optical system indicating the dierence, qualitatively, in the plate when iigured to correct for spherical aberration of the mirror object rays from the screen I3 must pass. The

cathode ray tube l0 may be of any suitable type and,'therefore, neednot be described in`detail,;-

The optical system comprising the mirror vIl fand the correcting plate I2 is of the same type as that described and claimed in the above-mentioned Landis application. The spherical mirror II by itself would project. an image having a, f

large amount of spherical aberration. Thaberration is removed by the correcting plate I2 which is properly shaped or iigured to give an image of high quality on the projection screen.

The shape Aof the particular correcting plate ilto the plane of the projection screen. As stated in the above-identified Landis application, the use of the spherical mirror and the correcting plate for correcting spherical aberration is based upon the principle of the well known Schmidt hef the tube envelope; This end of the ltube has a 'certain thickness of glass through which the i2," as gured in accordance'with my invention,

is represented in solid lines, while the outline of the correcting plate as would be figured without correcting'for the tube face thickness is indicated in dotted lines.

The fact that a substantial part of the total error in a system of thte above-described type is introduced by the end of the cathode ray tube was discovered fromv the following considerations. For the particular system described in this application (which isonehaving a 30-inch mirror) it is assumed that the maximum permissible ratio A ofthe diameter of the consequent circle of diffusion in the image to the diameter of the image is 0.002. Itis assumed that the axial position of the tube'is adjusted for best focus of the paraxial rays and'that ,th'excorrecting plate has been iigured to make the spherical aberration zero for zero thickness of the tube face. "Since itis assumed that an object without van intervening thick tube race is projected with perfect denni- 1- tion, the value A also represents the maximum permissible ratio of the diameter of the consequent circle of diffusion in the virtual image of the object produced by the tube face considered as' a lens to the diameter of the object.

Fig. 3 shows a light ray`traced from a point of the object which is on the axis, through the sin a=n sinv a-' y=t tan a Let p be the radius of the circle of diffusion due to spherical aberration on virtual image of the 3b-lect having a diameter d. Then a p=AI tan a A or A= 2 t(tan a sin a d "n' V ,1f-T- h2 a Solving for t for the case where d=7 inches c :aperture angle=4245' t :0.053 inch 2At tanga "ff:-

Since the projection tube of the size assumed should have an envelope about three times this thick at the large end or face to withstand the atmospheric pressure, it is now apparent that the circle of diffusion in the image will be too :large unless an additional correction is providedlin the optical system.

v One specific optical system ,designed in accordance with my invention will now be 'described by way of example. The method of designing this system will also be given, and, since the method applies to a system of any size, it will be clear to those skilled in the art how to apply the invention to optical systems having different lsize mirrors, different projection throws, etc.

f In this particular system, which is illustrated in Figure l, a front surface mirror inches in diameter is employed. A correcting plate diameter of three-fourths the mirror diameter was selected, since this results in even light distribution over the projection screen for the case where the central portion of the mirror is masked to\\ improve contrast as described and claimed in jection system.

length of the mirror are determined by deciding what maximum aperture angle a is permis-- sible from. the point of view of aberration in the system'. The angle which was adopted is 4245', corresponding to an f-number of 0.737. The corresponding radius of curvature of the mirror is! R=32.4 inches.

The size of the cathode ray tube iluorescent screen is such as to give, at least approximately, Vmaximum .brillianceof the projected image. A tube screen diameter d of 'I'inches was selected as being satisfactory.

The radiusof curvature r of the tube screen should be made approximately equal to, or slightf' ly greater than,onehali the radius of curvature of the mirror in order that the image will be sharp over the entire surface of a dat projection screen. A'curvature r of 16.7 inches was selected by a procedure which will be explained hereinafter. 'I'he type of glass for the tube face is Pyrex, with al mean index of refraction equal to 1.474. The thickness of the end of the tube or tube face is 1% inch.

The object and throw distances S and S', respectively, may now be determined by the approximate formulas:

where R is radius of mirror, m is the magnification, and f is the focal length of the correcting plate at itsI center or apex. In these formulas,

such 'parts in the path of a ray that are glass (i. e., the thickness of the cathode ray tube face and the correcting plate) are treated as equivalent to paths in air which are shorter in the in-vy plalned below, the value f=8.918, R=288.9 inches is selected.

November 29, 1940and entitled Reective pro- I The correcting plate is crown glass with an index of refraction equal to 1.532.

'I'he radius of curvature and, hence, the focal the optical system,. the value of S thus obtained would be an approximate vvalue and would be used as a check on other calculations. As explained hereinafter, the exact value of S is obtained by means of a more Aexact magnification formula..

With respect to the selection of the focal length ,f of the correcting plate, while an image which is perfectly sharp at the center can be obtained by the proper shaping of the correcting plate for any value of the central refracting power or focal length f ofthe correcting plate, the quality of the image off the axis isdependent on the choice of Here 'I'he ray is next refracted at the curvedV surface of the correcting plate indicated by subscript 3. The' radius of curvature of the cor- V"recting plate-at its center may now be found by the formula the value of f. In order to obtain the optimal value of f, numerous off-axis'rays should be calculated as will be explained hereinafter. Thus, by trial and error, the best value for f can be determined. This value of f is normally used in the where M is the magnification,

, (J'=8.918R), which corresponds to the object dis# tance thus determined and to the given magnifi-4 cation, is obtained by tracing a paraxial ray from Solving for r,

curvature and the focal length is:

of the correcting plateat its apex (and also the exact throw distance S). The value f will be Athe exact `focal length which will bring the rays yto focus at the desired throw distance. This ray tracing is done, as explained below, by means of the following Well known formulas for paraxial rays, in which s is the object distance and s' is n the image distance as illustrated in Fig. 4:

Reflection at a mirror of'radius R: V 1/si-l/s-i` 2/R Referring to Fig. 5.V (not to scale), a paraxial ray will now be traced by the formulas I through the system beginning with the center of the object at o. In order to illustrate the ray path, the ray in this figure is shown at a larger angle to the axis than a paraxial ray. Firsttracing the ray` through the glass face of the tube and into air, from the formula n/s'=n/s+(n''n)/r, we can iind si', r in this instance being the curvature of the outside surface of the tube face.

The ray is next reflected at the Vmirror.` For this reflection the object distance (si) equals the distance from the tube face to the mirrorplus s.

From the formula 1/s1=1/si+2/R, the subscript 1 referring to the mirror surface, we find that s'1= 873.9 (is negative since a paraxial ray diverges slightly after leaving the mirror).

The ray is next refracted at the plane surface of the correcting plate, which will be indicated by the subscript 2.

For lthis refraction, the object distance (sz) equals sfr plus `the distance from the mirror face ni particuiar, an the pcint h=o, wam- 0, 'as is usedin the process of obtaining the coefficients in the equation for the correcting plate curve.

The equation for the correcting plate curve is:

tu isV the thickness of the correcting plate at the axis and t is the thickness of the plate at the radial distance h from the axis. For the apertures here considered, lit is satisfactory to break oi the series beyond the term with ha. 'I'he slope of the corresponding plate curve is then given by v dt/dh=2ah+4bh3+6ch5+8dh The coeicients may be determined as fo11ows:' The first coeiiicient, 2a, is the reciprocal of the central radius of curvature of the correcting plate, or 2a=1/r.v This can be shown as follows: The reciprocal radius of curvature r of any curve, such l t, regarded as a function of the independent variable h [that is, of the curve t=t (h)-lis, according to a well known lawof analytic geometry, givenby L dit shown by substituting h=0 in the above equation 'forl dt/dh. l Hence, at this point, that is, at the vertex of the correcting plate curve,

A differentiation of the equation for dt/dh with respect to h yields for h=0 en, dhza Therefore As to is fixed to begin with, only three further coeiiicients, b', c and d, need be determined to fully establish the shape of the correcting plate.

The procedure for finding the coefficients b, c and' d is as follows: From five to seven rays leaving the center point of the object at a number of more or less evenly spaced angles to the axis are calculated through the system by the aid of the well known trigonometric formulas given below with reference to Fig. 6 (not to scale). These formulas are for rays in a meridional plane, that is, in a plane containing the sin 8= (n/n) sin 0 II s=stan/tan9 Refiecton at mirror:

`sin i= sin r sin i In the above Formulas II, the object and image distances are represented by s and s', respectively. The angle 0 is the angle of incidence of the ray with respect to the axis, angle 6 is the angle the refracted or reflected ray makes with respect to the axis, i is the angle of incidence of the ray at the surface under consideration, and the angle 'i' the vertex of the correcting plate by means of the formulas for refraction at a plane surface.

Having traced a number of rays by the Formulas II fromI the axial point of the object through the system up to the plane tangent to the vertex of the correcting plate curve, the height ofincidence h and angle of incidence 0, with respect to the axis, are obtained for each. It is then possible to find that slope of the plate surface which will just bend the incident ray so that it will. strike the axis at its intersection with the image plane. This slope is found as follows:

The throw distance S being known the anglelo of the refracted ray with the axis is given byv n application of Snells law to the case in quesis the angle of refraction or reflection at the surface in question.

Each of the several rays is traced throughgtheM system from the central point o of the object in steps similar to those described for a paraxial ray.

First considering the refraction caused by the tube face. the image distance s' and the angle 0' that the refracted ray makes with the axis are found by the formulas for refraction at a surface of radius r. The point at which the ray strikes the mirror is now found.

By the formulas for reflection at the mirror, the point at which the ray strikes the plane surface of the correcting plate is found.

The ray is next traced to the plane tangent tion then yields for the angle of incidence a the curved'surface of the plate:

sin (0"-0) tafn t1v-cos (0-0) I and the slope of the curve is given, finally. by

Table I shows the results for h and dt/dh for the rays calculated for the system being described.

Table I Anale of emer- Initial angle dI/dh Y (in gfazs) gentcb'om h (from tan van' 637' 1. 900585 0120150 8 l 1148 3. 371326 0199484 13 11)"17l 5. 4491K@ .0267316 Three of the slopes dt/dh thus calculated from tan pand the corresponding values of h are then substituted in turn'in the equation Thus the values of h and dt/dh substituted in the equation are those for the rays leaving the center point of the object and making 13, 22 and 2730'to the axis. Solving the resulting three simultaneous equations by determinants, the following values were found: 4b=5.08994.10-5; 6c=8.72839.108;

The depths of the curve as a function of h for this system is shown in Table II below:

The remaining pairs of values of Jt/dh and h are used to check t1 a correctness of the calculation and to determine the accuracy with which #usavo the curve represented by thepower series lits vthe ideal curve. The maximum deviation in slope which'has been found in any one case for the 7 2 2. 5" plate.is34 seconds of arc, corresponding to \circle of diffusion with a diameter equal to 0.0004Nof that ofthe picture.

If aconstant thickness to has been used in .tracing the rays through the system, a certain error will result; abetter approximation can be= obtained by deriving the thickness at the points of incidence of the several rays from the formula s with the values for b,.c and d just determined the iirst stagewas found to have a diameter of less 0.0003 ofA that of the picture. For plates with. much'steeper curves there mayv be an advantage 'in going to' 'the second approximation, but even here this can be avoidedif the thickness variation'is estimated from the proiiles of plates which have already been calculated.

lThe shape of 'the 22.5" correcting plate nally adopted is'foud to be given by It has sai-minimum' thickness of 0.57207 inch an" h=l0.0639 i'nch and curls up' slightly at the edge, Y'having there'(at h=ll.25 inch) a thickness of 0.58598 inch. .Y It has previously been stated that the best radius of curvature r for the tube face has been selectedto make the image fall in focus at all points on a flat projection screen `and that the focal length] of the correcting plate at its center is selected togive good marginal denition of the projected image as well as good definition at its center. The process by which these values r and f are selected will now be described more fully.

The procedure, when calculations are tlrst begun on 4a new system, is. to select values for r and f that look like reasonable values. The sys- -tem will then be calculated'using these values and the correcting plate curve is obtained. (This procedure has been described in detail, the de` scription being, of course, for a nal calculation with the nally selected values of r and f.)

The next 'step is to Vtrace-a plurality of off-axis rays through the system to iind outl howmuch they deviate horizontally and vertically from a principal ray leaving the same'oi-axs point on the object. The principal ray is represented by a straight line drawn from the said ofi-axis point through the centerof curvature of the spherical mirror.

Seven oi-axis rays ae traced inv the particular example being given, ve of them being in a vertical meridional plane (a meridional Plane being one in which lies the optical axis of the system), and two of them (which are referred to as skew rays) being at an angle witnrespect to the meridional plane and in a horizontal plane passing through the oli-axis point. The types of o-axis rays just referred to are illustrated in Fig. 'I where a plane surface is represented at P to aid in lillustrating the above-mentioned vertical and horizontal planes. .I By means of this ori-axis ray tracingf which L? IIormallyyfthis'l is not necessary: for' the 22.51

is done by means of the trigonometric formulas previously given and by means of the sk ew ray formulas given hereinafter.: there is found the horizontal and vertical distances separating the 5 point on the projection screen at which the offaxis ray falls and the point at which the principal rav'falls. After these deviation distances have been found for a system including the iirst calculated 1d correcting plate (having the rst selected value of f) and including .a tube with the tube face having the ilrst selected value of r, a. diierent value for r is selected and theseven ofi-axis rays are again traced through the system using the may now be plotted by drawing a straight line through each pair of deviation points obtained with the two values of r. It will be noted that an additional pair of skew ray curves are plotted for skew rays on the opposite side of the vertical plane and symmetrical with respect to the two skew rays traced through the system. From these curves, the value of the tube face radius'r which gives the minimum spread for a given cor- Yrecting plate of focal length f may be determined. As indicated by the vertical. arrows in' Fig. 8, this minimum spread for .the example plotted is obtained if the tube face radius 1' has a value of about 16.7 inches. It will be noted that this point o minimum spread is not necessarily thepoint either of minimum horizontal deviation or of minimum vertical deviation, but, s instead,-is a point where the deviation is the least, considering the two deviationstogether. The procedure for tracing the lve oi-axis rays in a meridional plane up to the plane surface ofA the correcting plate is the same as previously described for rays leaving the center of the ob-` ject` at a 'substantial angle,`the same trigono- 1w'metric formulas being used. The ray tracing from the curved surface to the projection screen is done"as follows:

Referring to Fig. 6 andto the angles indicated thereon for a. ray incident to the curved surface of the plate and refracted thereby, the ray strikes the curved surface at a distanceh from the axis which is found by the equationh=h2 +t2 tan 0 where 'ha is the distance from the 'axis to the point at which the ray strikes the ilat surface of the plate. and where ta is the thickness of the plate at this point and is also almost exactly the thickness of the plate at the -very slightly different and yet unknown distance h. J

The value of h thus found-is substituted in the equation y dt/dh=2ah+4bh+6ch5+sah1 to obtain' the velue 'of auchian e Then the angle 0 maybe calculated since imp-9 sin i=1iz sin i where -m is the indexof refraction of plate, and.

Y '=i' I I Finally the height of incidence H on the projection screen is -ltr- 41+[zr-Hinwil)1 ten o" AZISH-Ho Y where H is the height of incidence -of the prmcipal ray on the projecti\on screen. In tracing the two skew rays it is convenient sam'e f correcting plate. The 'curves of Fig. 8 'V a and car/e determined as follows:

, ttefraction at a spherical surface of radius r:

of f to obtain vsuflcient points to plot a curve of minimum spread vs. l/f, From this curve, there is selected the focal length f that gives the least spread, this being the value that is used in the final calculation of the sys-,

tem.

.There is also plotted a curve of optimum tube face radius 1' vs. l/f as shown in Fig. 10 from which is selected the best value of 1' for the value of f just chosen from the curve of Fig. 9.

1' It will be seen that, by the above-described consideration of ofi-axis "rays, vthe optimurr' values have been chosen for the tube face radius r and for the focal length f of the center of'the correcting plate, whereby the system will project aberration, this type of aberration being small as compared with the geometrical aberrations con.-

an image of excellent quality at the margin as well as at the center of the image.

It may be noted that the .system which has been described is well corrected vfor chromatic.

sidered in the foregoing calculations. With respect to the shape or curve of the cor- Refraction at alplane:

a, a'im'mr-l) 1+a2+c2 +111f2 where It will be apparent that, by means of the foregoing equations, a skew ray may b e traced from point to point through the system to determine the point at which it falls on `the projection screen.- It is then a simple matter to calculate the distance horizontally and the distance vertically that this point is spaced from a corresponding principal ray.

recting-plate when it is figured in accordance" with my invention as compared with its shape when figured without taking into account the thickness of the tube face, it has been pointed out with reference to Fig. 2 that thecorrecting plate in the improved system is relatively less concave at the edge. This diierence in the corrected and uncorrected curve of the correcting plate may be expressed mathematically as ex'- plained below. The correcting plate curve with- 40 out tube face correction may be defined as fol-v where the distance s' and the angles i9l and 0' are those shown in Fig. 6 atthe tube face and where n is the index of refraction of the tube face.

Let At=differencein thickness of a correcting plate which is not corrected for the' thickness of the tube face and of a correcting plate which .is so corrected. The ray leaving the center point of the object at the angle 0 will, continued backward after Having traced the oli-axis rays` through the system for onecorrection plate of focal length f and for two values of tube face radii 1' whereby a set of curves such as those in Fig. 8 is obtained, the next step is to select another value of focal length f for another correcting plate and repeat the procedure; again tracing through a number of olf-axis rays. Thus, there is obtained another group of curves similar to those in Fig. 8, but for a different f value.

For this second f value, the tube face radius 1 that will give minimum spread is again selected. This procedure is repeated for several values refraction at the tube face, intersect the virtualA i image plane ata distance om the,axis. It, as with the correcting plate uncorrected for the thickness of the tube face,

the virtual object plane is imaged sharply on the projection screen, the actual ray leaving the object at the angle 0 will strike the projection screen a distance 5 from the axis, m being the magnification. Thus, i

as shown in Fig. 9.

geteste l to correct the plate for the tube face thiol-ness its slope at the height of incidence o1 the ray considered must be changed by an amount d: AIE v such that the angle oi refraction at the correcmg plate is increased by an amount i. e. jest enough to cause the ray considerec to As, in order toproduce an angular deecaflon As' the slope ci the reiracting surface has te be changed by an angie 1 dt dat such that A6' A A=nz-1 the needed change in the slope of the correctirg plate is where S' is the throw distance and nz is the irex of refraction of the correcting plate` As A di -2s Yf 1 cos 6' Where R is the radius of curvature ci the spherical mirror. Again, to a ifirst approximation where F is the focal length of the system We have now found 'now the coecient la, c and d in the equation for the correcting peste The numerical values for Ab.. Ac and Ad are given below for the particular projection unit described in this application,

ca -275.0543 A6- "Lirleosztsaof-m lo 7 L 18m10.516s ad meaeszsanf-7840- v Correcte Uncorrected "c: 1.2725-10-5 {1.293540-5) e iessrsw (Lessors-2) d: 5,5916-:101i (Meerlo-11) I claim as my invention:

l. A projection system comprising a concave spherical mirror, a correcting plate positioned at least appro. ately at the center or' curvature of mirror. and a projection tube having an objc-:t surface therein which is positioned at a coniugate focus of the system and having a trans- 5 parent refractive end or tube face of a certain 1 24e Q tthickness, said correcting plate being gured to correct both for the spherical aberration of the mirror and for the error introduced by said thickness of trie tube face.

2, A projection system comprising a concave spherical mirror, a correcting plate positioned at least approximately at the center of curvature of said mirror, md a cathode :ay tn'ee having a luminescent screen therein which' is positioned at a conjugate focus of the system and having a transparent refractive end or tube face of a certain Cxickness, said correcting plate being gured to correct both for the spherical aberration of the mirror and for the spherical aberration introduced by said thickness of the tube face.

3. A projection system comprising a concave 1 sos K 1 (111-, h5 1-p f l0 1 -nrW/h spherical mirror. a correcting plate positioned at least appremmateiy at the center o crr'v'atuse of coniugate focus of the system and hnving s transv parent refractive end or tube fece ci certain thickness, said correcting plate being gured to correct for the spnericai aberration o! the mirror and for the edditiona error introduced by the refraction caused by the end of said tube whereby said plate relatively leas concave on the edge than a correcting plate figured tc correct only for the spherical aberration of the mirror.

4. A projection system comprising a concave spherical mirror, a correcting plate positioned et least approximately at the center of curvature of said mirror, and a projection tube having an object surface therein which is positioned et e. coniugate focus of the system and having a t transparent refractive end or tube ieee ci a certain thickness, said correcting plete being figured to correct both for the spherical aberration o the mirror and -or the error introduced by said thickness of the tube fece, the correcting plate curve being substantiaily defined by the equation escasos l. end or tuoe face of substantiel thickness through which the light rays from seid 'iight image pass, and encpiical system including a substantisily spherical mirror having its concave surface area positioned to receive said light rays from the iml ege, and an aspherical zone plete positioned ex termal to the path of the light projected from the light image to the reflector and located at or near the center of curvature of said mirror to receive the reected light from the reflector, said zone plate having such curvature as to correct both for the spherical aberration introduced by the rellector and for the spherical aberration introduced 'oy said thlclrne of the tube face,

where-t is the thickness of the correcting plate et distance h, h ic the :radial distance meee ured from the center of the correcting piste, tu is 35 Less of the toire fece, n is the index of refraction 5,5

of the tube face, m is the index ci refraction of the correcting plate, and R is the radius of curveiture of the spherical mirror.

5. im image projection device comprising e. cathode ray tube having a bidimensional imsge eren on which there may be produced a light image, said tube having a. transparent refractive end or tube face oi substantiel thickness through which the light rays from said light image pass, and aiu optical system including o concave reectng surface of revolution having its conve surface avea. positioned to receive seid light rays from the image, and an aspnerical zone plate positioned external to the path of the light projected from the iight image to the reiiector and positioned to receive the reected light from the reector, seid zone plate having such curvature as to correct both for the spherical aberration introduced by the reflector and for the sphericei aberration introduced by seid thickness r whereby the optical system including the reflector and tire sone plete is adapted to form a projected spherical mirror having its concave surface aree.

positioned to receive said light rays from the imese, and an aspirericei zone plate positioned extel-nal to the path of the light projected from the ghi image to the reector and located at or near the center of curvature of said mirror to receive the reflected light from the reflector, said zone,

curvature also being such as to reduce the seid spheril aber-intiem to a minimum in n vievv'ir-ty area located at e, finite distance from the said zone plate.

EDWARD G. RAEx/IEE'G

Images