US2474288A - Octantal error calculator - Google Patents

Description

June 28, 1949.

Filed July 7, 1945 G. SWIFT ETAL OCTANTAL ERROR CALCULATOR 3 Sheets-Sheet 1 FIG.1.

INVENTORS. GILBERT SWIFT BY CHARLES B. MOORE ATTORNEY 6'. SWIFT ETAL OCTANTAL ERROR CALCULATOR 'June 28, 1949.

3 Sheets-Sheet 2 Filed July 7, 1945 000 00B 000 00V 000 oON oo 00 INVENTORS. GILBERT SWIFT BY CHARLES B. MOORE QAZL-MQM ATTORNEY G. SWIFT ET AL OCTANTAL ERROR CALCULATOR June 28, 1949. v

I s Sheets-Sheet 3 Filed July 7, 1.945

INVENTORS. cu. BE RT sw F CHARLES B. RYE- ATTORNEY Patented June 28, 1949 UNITED STATES OFFICE OCTANTAL ERROR CALCULATOR Gilbert Swift, Tulsa, Okla., and Charles B. Moore, Elberon, N. J.

Application July 7, 1945, Serial No. 603,738

10 Claims.

The invention described herein may be manufactured and used by or for the Government for governmental purposes, without the payment to us of any royalty thereon.

The present invention relates to calculating devices. The invention finds its primary application in connection with radio direction finders, some types of which have inherent but calculable errors. The principles may however be applied to other cases where a. similar problem is involved. The calculating device as applied to direction finders is used to convert observed bearing angles into true bearings without the necessity of referring to error tables or other data.

In direction finders of the spaced antenna type, such for example, as that of Adcock, the resultant induced voltages follow accurately the well known cosine law, for varying angles of arrival, only when the spacing between the antennae is small relative to the wave length of the incident radiaion.

Direction finders operating at moderately high frequencies of the order of say to 30 megacycles, have in practice an antenna spacing which is an appreciable fraction of a wave length. The resulting antenna induced voltages consequently do not follow the cosine law accurately, so that when used in conjunction with a well designed radio goniometer which does follow the true cosine law, errors result. This type of error may be calculated from the known data of antenna spacing and frequency used, and is known in the art as octantal error since the values repeat themselves every 45 degrees. These calculated errors may be prepared in the form of tables for each bearing angle and frequency or may be plotted in the form of curves.

The present invention avoids the necessity for having to refer to prepared tables or curves of octantal error by providing this data in a more immediately usable form. This data is so plotted on the face of a circular dial forming part of the calculator, that it is only necessary to set the pointer at the observed bearing angle and then read off the true bearing for the frequency being used.

In addition to octantal error, there exists in direction finders of the type named another kind of error known as zero shift which is not calculable and varies irregularly with the frequency and the special that is, local conditions of the installation. This kind of error can only be determined by experiment and is usually recorded in a table giving the error for the various frequencies. The

movable arm of the calculator is arranged so that corrections for zero shift may be plotted in such a way that corrections for both octantal error and zero shift are combined into one operation.

It is an object of the invention to provide a calculating device which will facilitate the work of deriving true bearing angles from observed bearings which require to be corrected for octantal error.

It is a further object to provide a calculating device which will combine in one operation the corrections for zero shift and octantal error.

The invention will be understood by reference to the drawings in which,

Fig. l is a general view of the octantal error calculator, embodying the invention,

Fig. 2 is a view of the pointer or movable arm of the calculator of Fig. 1 on an enlarged scale,

Fig. 3 shows one method of plotting calculated octantal error data.

Fig. 4 shows the method of plotting the octantal error data of Fig. 3 on the circular dial of the calculator, and

Fig. 5 is a view showing a graph on which corrections for zero shift errors are plotted and transferred to the calculator arm.

Referring now to Fig. 1, the circular dial 22 is provided with an inner graduated circle 24 for measuring indicated bearing angles. In a zone between the intermediate and outer circles 28 and 30 a series of curves 32 are plotted. These curves are identified by the large reference numerals 34 in a scale of degrees at the outer rim of the dial and by the intermediate graduations 36, which represent true bearings in degrees and these curves are drawn by methods to be described later.

The movable arm or pointer member 40 is pivoted conventionally to the dial center 26. An index mark 42 is on the arm centerline and registers with the graduated circle 24 at the inner edge of a window opening 10 in the arm 40. The index center line is continued in part through the outer part of the arm as shown at 44. An irregular index edge 48 is derived from experimental data giving corrections for zero shift in a manher to be described. That part of the arm between dial circles 28 and 30 is provided with graduations 46 each representing a respective frequency in megacycles. These graduations are shown on the drawing as equally spaced, but equal spacing is not essential, in fact the dial curves may be made more nearly straight by crowding the frequency graduations toward the outer rim of the dial.

The dial may be made of any suitable material on which the data may be printed or engraved 3 such as metal, Celluloid or even heavy paper. The pointer may be made of similar material but is preferably made of some transparent plastic such as Celluloid, etc., on which the center line, index mark and frequency graduations are engraved.

The calculator dial is fixed and permanent for the particular antenna spacing and frequency range for which it is designed. When the calculator is intended for octantal error corrections only, the pointer arm is provided with a centerline, index mark and frequency graduations and when assembled on the dial the instrument is complete.

The instrument is however, preferably arranged to include corrections for zero shift which requires the irregular line 48 to be plotted on the pointer arm. Since this correction varies from installation to installation it is customary to provide a number of pointer arms with each instrument. A new arm properly corrected for zero shift is pre-- pared from experimental data for each set up.

Values of octantal error, calculated by methods well known in the art, are usually plotted in curves with error as ordinates and bearing angles as abscissae. This data may be put in a more convenient form when plotted as in Fig. 3 where true bearings are plotted vertically against indicated bearings horizontally. True bearings are derived from indicated bearings by adding the calculated correction for octantal error. Curves are plotted for each of the several frequencies as shown by way of example. The data in this form however, is not suitable for plotting on the calculator dial.

The curves shown on the calculator dial, Figure 1, are derived from the curves of Figure 3 by the method shown in Figure 4. A series of circles are described about the dial center 26 and are numbered 5, 6, 8, 28 which correspond to the same frequencies in megacycles shown in Figure 3. A series of equally spaced radial lines are drawn through the center 26 and marked with the appropriate bearing angle in degrees measured clockwise from the vertical line marked il.

As an example, consider the horizontal abscissamarked (Fig. 3), all points of which correspond to a true bearing angle of 10. To plot this curve follow the 10 true bearing line of Figure 3, horizontally until it intersects the first frequency curve at the point 59. Read off the indicated bearing angle 105 on the horizontal scale and note that the frequency for that curve is 6 me. This point is then plotted on the dial Figure 4 by following the 6 mc. circle through an angle of 105 to the point Ell. The next curve intersected by the 10 line, Figure 3, gives the point 52 on the 8 mc. curve at the indicated bearing of 11.5 and is plotted on the 8 mo. curve, Figure 4, at the bearing angle l1.5. This process is continued point by point as we follow the 10 curve of Figure 3 through its successive intersections with the constant frequency curves, the final point '64 occurring at 27 on the mo. curve and is so plotted in Figure 4.

The 20 curve is plotted in a similar manner by following the 20 true bearing line of Figure 3 horizontally through its intersections with the constant frequency curves and plotting the corresponding indicated bearings in Figure 4 as before.

In this way a series of constant true bearing curves at 2 intervals may be built up as shown in Figure 1.

It is a property of this system of curves that the intersection of the pointer centerline with any particular dial curve gives the frequency for which that curve gives the true bearing. As an example consider the arbitrary position of the pointer arm shown in Figure 1. The pointer index is set at 35' which is the indicated hearing as given by the direction finder. The pointer centerline 64 intersects the 30 true bearing curve at 12.5 megacycles as read on the frequency scale marked on the arm. In other words, the true bearing corrected for octantal error (but not for zero shift) corresponding to the indicated hearing of 35 and frequency of 12.5 mc. is 30 If zero shift is neglected or corrected for sep arately, the calculator as so far described is complete and provides adequately for octantal error corrections. True bearings are read with the aid of the simple centerline M marked on the arm, if made of transparent material such as Celluloid, or if opaque, the arm may be cut to the centerline as an edge and values read from this edge.

The instrument is however, preferably arranged to include means for correcting for zero shift. This accomplished by plotting the corrections on the pointer arm in the form of an irregular curve such as 48 shown in Figs. 1, 2 and 5.

The method of adding the correction for zero shift is as follows. A number of pieces of graph paper as shown in Figure 5 are supplied with each instrument. The radial lines are drawn at one degree intervals for a distance of 10 on either side of the 0 line. The concentric lines of the scale on the arm 40 marked 5 to 20 represent megacycles of those frequencies, consecutively from the outer part inward and correspond accurately to the correction value indicated where these frequency markings on the pointer arm intersect with shift lines 32, Figure 1. A table of zero shift errors is compiled from local experimental data for the required frequency range. These corrections are plotted on the graph paper. Thus, for 8 mc., if the error is .1, place a point at .4 (to the left of the zero line) on the 8 mo. concentric line. If at 15' me. the zero shift is 2.8, place a point at -2.8 on the 15 mc. concentric line. In this way, the line 48 is plotted. After the graph has been plotted, place it over the pointer arm Figure 2 so that the 0 radial line coincides with the pointer centerline and the 5 mc. and 20 me. concentric lines on the graph coincide with those on the arm. Holding the graph paper securely in position, transfer the points previously placed on the graph paper to the arm by pricking the arm through the points on the graph with any sharp pointed instrument such as a needle or awl. After the points have been transferred to the arm, remove the graph paper and connect the points on the arm with straight lines cut in the Celluloid by an awl or scriber.

When the pointer arm is made of transparent material it may be assembled on the dial and used without further alteration. If, however, the material is opaque, it is necessary to cut away that portion included between the dotted line '58 and the irregular line 48. It is also necessary to cut an opening or window 10 in the arm so as to expose the graduated dial circle 24 with which the pointer index 42 registers. These alterations may be made, whenpreferred even when the pointer arm is transparent and are so shown in Figure 1.

The completed calculator provided with corrections for zero shift is used as follows. Rotate the pointer arm until the index 42 registers with the indicated bearing on the dial. Hold the arm in this position and follow' out on the centerline to the concentric line corresponding to the frequency of the received signal. Follow this concentric line to the edge 48 that has been cut to compensate for zero shift. From this point. fol.- low out on or between the curved lines on the dial to the scale on the outer edge. Read the true bearing in degrees on this scale where the curved line intersects. The result is the true bearing corrected for octantal error and zero shift.

'It will be noticed that the constant bearing curves of Figure 4 converge toward the point 65 for angles between 30 and 60. The use of the calculator is inadvisable in the vicinity of these points of convergency, which recur at four points on the dial, since in these regions a slight error in the observed bearing results in a large error of the corrected bearing. To avoid improper use of the calculator for settings in these undesirable regions, these localities are blotted out as shown at l2.

Having described the invention, what is claimed is:

1. A calculating device for indicating true azimuthal angles in radio direction receiving apparatus for radio-waves having deflection errors, comprising a dial provided with a graduated circle for measuring indicated bearing angles, curves on said dial having angular variation progressively and proportional to true constant bearings for corresponding indicated bearing correct ons and frequency; means identifying the bearing values of said curves; a pointer arm pivot; iy mounted at the center of said dial, said pointer having a centerline and an index mark at said circle and radially spaced graduations corresponding to respective frequency values at their intersections with said dial curves.

2. A calculating device for indicating true azimuthal angles in radio direction receiving anparatus for radio waves having deflection errors,

comprising a dial provided with a graduated circle for measuring indicated bearing angles, curves on said dial corrected for octantal error proportional to true constant bearing corrections for corresponding indicated bearings and frequency; means identifying the bearing values of said curves, a pointer arm pivotally mounted at the center of said dial, said pointer having a centerline and an index mark at said circle and radially spaced graduations corresponding to respective frequency values at their intersections with said dial curves.

3. A calculating device for indicating true azimuthal angles in radio direction receiving apparatus for radio waves having deflection errors, comprising a, dial having a first graduated circle for measuring indicated bearings, a second circle having true bearing graduations thereon, true constant bearing curves terminating at outer ends on said second circle and having angular variation progressively inward on the dial proportional to ascending frequency values and accompanying deviations; a pivotally mounted pointer member having a centerline, an index mark and radially spaced frequency graduations corresponding to respective frequency values at their intersections with said dial curves.

4. A calculating device for converting indicated bearing angles of radio direction finders into true bearing angles corrected for octantal error according to frequency comprising a dial having a first calibrated circle for measuring indicated bearings, a second graduated circle, true constant bearing curves terminating on said second circle and having angular variation progressively in a radial direction proportional to ascending frequency values and deviation corrections therefor; a pointer member pivotally attached to the center of said dial, said pointer having radially spaced frequency graduations corresponding to respective frequency values at coincidence with said dial curves and having a centerline and an index mark at said first circle, whereby points on said true constant bearing curves having respective correction values will be indicated an said center line at respective radial distances to show a bearing in-- cluding correction for octantal error corresponding to the frequency graduations on said pointer arm.

5. A calculating device for correcting radio direction finder bearings comprising a dial having a graduated circle for measuring indicated bearing angles, a second graduated circle, true constant bearing curves terminating on said second circle and having angular variation propressively proportional to ascending frequency values and related deviations; a pointer member pivotally attached to the center of said dial, said pointer member having a centerline, an index mark and equally spaced frequency graduations; said true bearing curves representing plotted curves of constant true bearing points at respective indicated bearing angles measured on said pointer arm at said frequency graduations substantially as described.

6. A calculating device for correcting indicated radio bearings for both octantal error and zero shift comprising a dial and a pointer arm pivotally mounted on the dial, said dial having a first graduated circle for measuring indicated bearings, a second graduated circle for true bearings, true constant bearing curves corrected for octantal error terminating on said second circle and having angular variation progressively from said circle in proportion to ascending frequency values and deviations; said pointer member having radially spaced frequency graduations corresponding to angular deviation of waves of respective fre-- quencies indicated at similar radial points on the dial curves, said pointer having a centerline, an index mark at said first circle and a zero shift correction curve, points on said curve being displaced laterally on either side of said centerline by an amount equal to a curve of correction for zero shift.

'7. A calculating device for correcting indicated bearing of a source of very high frequency and ultra high frequency radiant energy for both octantal error and zero shift comprising a dial and a pointer arm pivotally mounted on the dial, said dial having a first graduated circle for measuring indicated bearings, a second graduated circle for true bearings, true constant bearing curves corrected for octantal error extending from said second circle toward the first said circle; said pointer member having radially spaced frequency graduations corresponding to values at corresponding points on the dial curves, a centerline on the arm, an index mark at the first circle and a zero shift correction curve, on the pointer the material of said arm being cut away along the line of said curve to form a reading edge, points on said curve being displaced laterally on either side of said centerline by an amount equal to the correction for zero shift.

8. A calculating device comprising a surface of revolution having an axis, pointer means adjacent to said surface having an indicated angle 7 index and spaced frequency indices thereon, said surface and pointer being relatively movable about said axis, a graduated scale on said surface to measure indicated bearing angle, and a series of curves plotted on said surface indentified by true bearing angles and cooperating with the frequency indices to indicate the true constant bearing corresponding to the indicated bearing for the various frequency values.

97 A device for correcting octantal error in radio direction finders receiving waves from spaced antenna, comprising a plate having a circular dial thereon comprising a circle calibrated in azimuthal units, a circular zone concentric with said circle having quadrate groups of curved lines terminating at the periphery of the zone spaced there in uniform azimuthal units and varying angularly and progressively from a cardinal line dividing the groups, and being convergent mutually and generally toward a medial radial line in each group, and a pointer pivoted coaxially of the circle having an index associated with said circle and including a coincident index line extending across said zone, and having radially spaced index marks extending to said index line to indicate points on and in relation to said curved lines, angular variation of said curved lines being proportional to a curve of error in direction of a radio beam according to its frequency and direction with respect to each said group, and said index marks being spaced radially in proportion to respective radio frequencies, whereby junction of any one with one of said curved lines at said index line will indicate the azimuthal correction at the said termination of the intersected cur \ved line.

10. The structure of claim 9 wherein said pointer is formed in addition with a curved index line extending generally from its outer part coincident with said index line inward and divergently therefrom progressively and in numerical proportion to a local zero shift error whereby intersection of said curved index line with any of said first named curved lines at said index marks will indicate the further azimuthal correction to compensate for zero shift at the particular device.

GILBERT SWIFT. CHARLES B. MOORE.

REFERENCES CITED FOREIGN PATENTS Country Date France May 13, 1932 Number

Images