DE1132365B - Analog-digital converter for converting a rotation angle value into a binary numerical value using a rotary switch - Google Patents

Description

GERMAN

PATENT OFFICE

T15121 IXc / 42m

REGISTRATION DATE: MAY 6, 1958

NOTICE
THE REGISTRATION
AND ISSUE OF
EDITORIAL: JUNE 28, 1962

The invention relates to an analog-digital converter for converting a rotation angle value into a binary numerical value by means of a rotary switch.

Digit calculators find the solution because of their high accuracy and adaptability various tasks in modern science and technology an ever-increasing application possibility. With the automatic control of Production, measurement and other processes, however, encounter the full utilization of such calculating machines difficulties, because the conversion of analog, continuously changing quantities into corresponding ones Numerical values are not given with sufficient accuracy.

Previously known methods require precisely divided linear or circular scales, which is also the case Applying the most accurate production methods the final conversion accuracy to about 0.03% (fifteen binary digits). Thereby the manufacture is one with such accuracy working device is very expensive, and the device itself is very large, the minimum length its scale is 1000 mm.

Known converters mainly use two to avoid ambiguity in the transition Methods:

a) output from the converter in symmetrical code (gray code, cyclic code),

b) the converter has two to four brushes that can be switched with electron tubes or relays depending on Magnitude.

The solution according to a) is disadvantageous because it is necessary to have an additional decoder for the stationary Connect a converter for converting a number into a normal binary code. The solution b) requires more electron tubes, transistors or relays for the decoder, as many as that Converter should have binary digits or, as with a known converter, a double number of Diodes.

Attempts have already been made to resolve the ambiguities in the known transducers remove. Photoelectric reading transducers have been used for this purpose.

A known converter has an output with the direct binary code, in which to Elimination of errors when using additional elements. Other known converters have a relay for each order of magnitude or a relay, a transistor and a diode for each order of magnitude or four brushes and two diodes built into the converter of each size.

Analog-digital converter for converting a rotation angle value into a binary one Numerical value by means of a rotary switch

Applicant: TESLA, närodni podnik, Prague

Representative: Dipl.-Ing. A. Spreer, patent attorney, Göttingen, Groner Str. 37

Zdenek Fixa,

Vsenory, Dobrichovice (Czechoslovakia), has been named as the inventor

None of the known converters has a direct output without additional elements such as relays, electron tubes or diodes, etc. The number of these elements is

as relatively high, one to two elements per order of magnitude.

The essence of the invention, which focuses on an analog-digital converter to convert the values an angle of rotation into a numerical value expressed in the dual system using a rotary switch relates, is that the size of the input angle to be scanned directly or via conversion gear affects the angular position of the brush of the first rotary switch, followed by a series of rotary switches, each of which has two brushes with an even number of mutually insulated conductive segments work together and are arranged one behind the other in such a way that the number of switched Segments of each individual switch in a unit of time is twice as large as the number of switched Segments of the immediately following switch, with two segments of each A positive or negative voltage is connected to the switch and to two more with these segments adjacent segments the output and inverse output of the immediately preceding switch is connected and the brushes of each switch, which are adjusted against each other by two segments, the output or the inverse output from this switch in the direction of the immediately following one Mediate switch.

209 617/210

Each of these rotary switches corresponds to a binary digit of the desired output number. By using twenty rotary switches of this type, a twenty-digit dual number can be achieved, which corresponds to an accuracy higher than ΙΟ " ⁶ (0.001% o) for the entire measuring range Convector wave can be increased at will.

Segment, etc. One of the segments is connected to the output of the switch of magnitude n, -1. A negative voltage is connected to the segment immediately following in the positive direction indicated by the arrow. A positive voltage is connected to the immediately preceding segment, and to the segment preceding this segment the inverse output from the switch is of the order of magnitude η - 1

Berm transition from a lower to a 10 connected. To the brushes, the middle one

The higher numerical value was obtained from the following in the design of the transducer according to the invention Properties of the dual numbers made use of:

^{1. If the 1} S value changes in the order of magnitude (n - 1) from 0 to 1, then the value in the order of magnitude η does not change.

a) With every odd-numbered change, the value remains in the order of magnitude «equal to 0.

b) For every even-numbered change, the value remains in the order of magnitude «equal to 1.

2. Changes in the order of magnitude («- 1) of the Value from 1 to 0, the value changes at the same time in the order of magnitude «.

a) For every odd-numbered change, the state of the order η is inversely related to the state of the order («- 1).

b) For every even-numbered change, the state of the order of η is identical to the state of the order of (n - 1).

35

The invention will now be described in more detail with reference to the exemplary embodiment shown in the drawing be explained.

Fig. 1 shows a four-segment rotary switch according to the invention of the order of magnitude;

FIG. 2 also shows a rotary switch of the order of magnitude, but with sixteen segments;

3 shows, for example, a circuit according to the invention with a plurality of rotary switches.

As can be seen from FIG. 1, the output of the rotary switch becomes of the order of "-1 one of the segments of the rotary switch of the order of magnitude «and the inverse output from the Switch of the order of magnitude «- 1 the opposite segment of the switch of the order of magnitude «Fed. A positive or negative voltage is connected to the remaining two segments. The output from the switch of the order of magnitude «, which is the switch of the immediately higher Order of magnitude «+ 1 is supplied to one brush, while to the second brush this switch is connected to the inverse output of the order of magnitude «. The positive direction of rotation is indicated by the arrow.

Fig. 2 shows that the number of segments of the rotary switch can be a multiple of four. The illustrated switch of the order of magnitude «has sixteen segments. The single segments of the Switches are always switched through over four segments, i. H. the first, fifth, ninth, and thirteenth Segment; the second, sixth, tenth and fourteenth angles double the spacing of the segments equals, the output or the inverse output from the switch is of the order of magnitude «in direction connected to the switch of the order η + 1.

The mutual mechanical connection of the individual switches is designed such that the Number of switched segments of the switch is of the order of «twice greater than that Switch «+ 1. This can be done either by using switches with the same number of Segments and a speed ratio of 1: 2 between adjacent switches or by direct mechanical connection between neighboring switches, where the number of the segments has a ratio of 2: 1. Also, these two solutions can be based on can be combined with each other in different ways.

A series of such mechanically and electrically connected switches enables the conversion a value given by the rotation of a shaft into a dual number, this series being replaced by a first switch and a resistor network according to FIG. 3 is supplemented in such a way that the change in state of the direct and inverse output from the switch of the zeroth order of magnitude exactly at the same time he follows.

The converter shown in Fig. 3 operates as follows:

The switches F ₀ , V \, V ₂ etc. are mechanically connected to one another by gear ratios that do not allow slipping in such a way that two turns of switch F ₃ , four turns of switch F ₂ , eight turns of switch F ₁ etc. one turn of the switch F _{4 is} omitted.

The switch F ₀ forms a simple contact which is connected between the points χ and y of the resistance bridge arranged between the switches Vq and Fj. This bridge is fed with a direct voltage, possibly also with an alternating voltage ES, which is symmetrical to ground. The resistances, B of the bridge are selected such that when the switch Vq is open, there is a voltage of size e between points u and ν of the bridge, the polarity or phase of which is reversed with respect to the polarity or phase of the voltage E.

This state expresses that the digit 0 is in the zeroth order of magnitude of the number. When the switch F ₀ is closed, i.e. when there is a short circuit between points χ and y, there is a voltage of the same magnitude e between the starting points u and ν, but with the same polarity (phase) as the voltage E. This property of the bridge can be achieved if the values, B and C satisfy the following equation:

2 B ² - C2 + 2 AB + AC + BC = 0.

(For example A = 20 kß; B = 50 kß; C = 126 kß.) The value of the output voltage is then

"C CAB

C + A + B

(in the example e = £ 0.286). A voltage of this value must be applied to the points which are denoted by + and - in FIGS. 1 and 2. In order not to have to provide a special voltage source, three further resistors are connected to the six resistors that make up the bridge, which form a resistance voltage divider D, F, D. As a result, this voltage e is derived from the supply voltage E.

This voltage divider is calculated so that the following relationship holds

.F _ CAB

2D + F C + A + B

a part of a transducer that has an angular position of the shaft using a seven-digit binary number with an accuracy of about 2.8 °.

By connecting a further, similarly arranged group, the brushes of which, however, depend on the shaft revolutions in a gear ratio of 1: 64, an exact specification of revolutions from 0 to 63 can be achieved. The rotary switch V ₀ is not used.

ίο The brushes of the rotary switch V _{6 of} the first group are connected to the segments of the rotary switch V \ of the second group instead of the starting points u, ν of the bridge, which is omitted in the second group.

The groups of the second type can be added at will, and this also becomes the area of the converter extended to higher speeds.

(in the example under consideration, one chooses e.g. F = 64 kü, D = 80 kfl).

The binary output number can then be tapped, either from the outputs 0, 1, 2 etc., where a negative voltage value expresses the number 0 in the relevant order of magnitude and a positive voltage value expresses the number 1, or from the outputs 0, 1, 2 etc. . in inverse form.

A change in the input variable, ie a change in the angle of a shaft, a lever mechanism, etc., influences the change in angle of the brush of the first rotary switch V ₀ , either directly or via a gearing. The purpose of this rotary switch V ₀ is to secure the changeover of its brush with segment 0 or segment 1. Each of these segments can take up either 180 ° or any whole part of 180 ° on the switch. The subdivision of the individual rotary switches depends on the mechanical construction and is usually chosen as a whole power of 2, 2 ¹ , 2 ² , 2 ^{3 etc. One can choose 2 6} = 64 for the zeroth order of magnitude. This creates a converter that converts the position of the input shaft into a number with an accuracy of 1: 128 of the revolution. It is therefore possible to scan a transferring angle in numerical form with an accuracy of 2.8 °. A finer subdivision could be achieved by a special construction, e.g. B. by increasing the diameter of the rotary switch, or reducing the brush dimensions od. Like. Can be achieved.

If a rotary switch with one hundred and twenty-eight contacts is used for the zeroth order of magnitude, then the rotary switch of the first order must also have one hundred and twenty-eight contacts in the case of direct connection with the rotary switch. These contacts are switched through in a manner similar to that in FIG. 2, where only sixteen contacts are shown for the sake of simplicity. The further rotary switch V ₂ is switched through in the same way, but the number of changeover contacts is reduced to sixty-four. F ₃ is provided with thirty-two contacts, F ₄ with sixteen contacts, etc. to V ₆ with four contacts. The rotary switches F ₀ to F ₆ become a mechanical unit, e.g. B. formed by printed circuits, the brushes are controlled directly as a function of the input angle. This arrangement of the rotary switches V ₀ to F ₆ forms

Claims (5)

PATENT CLAIMS: 1. Analog-to-digital converter for converting an angle of rotation value into a binary numerical value by means of a rotary switch, characterized in that the size of the input angle to be scanned directly or via conversion gear influences the angular position of the brush of the first rotary switch, which is followed by a series of rotary switches, each of which has two brushes that work together with an even number of mutually insulated conductive segments and are arranged one behind the other in such a way that the number of switched segments of each individual switch in a unit of time is twice as large as the number of switched segments of the immediately following switch, that a positive or negative voltage is connected to two segments of each switch and the output and inverse output of the immediately preceding switch are connected to two further segments adjacent to these segments and the brushes of each switch, which are mutually adjusted by two segments, convey the output or the inverse output from this switch in the direction of the immediately following switch. 2. Analog-digital converter according to claim 1, characterized in that the rotary switch with the same number of segments are connected in series by a mechanical translation in such a way that the rotational speed of the arms of each switch is equal to half the speed of rotation of the arms of the immediately preceding switch. 3. Analog-digital converter according to claim 1, characterized in that the arms of all Rotary switches rotate at the same speed and that each switch doubles Has number of segments of the immediately following switch. 4. Analog-to-digital converter na -. 'H claim 1 to 3, characterized in that £ Jar rotary switch of the lowest order of magnitude, whichever has only one arm, with the rotary switch of the next higher order of magnitude is connected via a resistor network, which is designed as an unbalanced bridge, and the first rotary switch is connected between two symmetrical points, the two Divide opposite branches of the bridge into two unequal parts. 5. Analog-digital converter according to claim 1 to 4, characterized in that one of each two opposing failures of the bridge on each of two opposing segments of the immediately following rotary switch is connected to the other two At the top of the bridge a voltage divider is connected, from which the two other segments the voltage supplied to the switch is derived. Considered publications:
"The Institution of Electrical Engineers", Vol. 103, Part B, pp. 42 to 51, no. 7, January 1956;
"Instruments and Automation", Vol. 29, H. 6, June 1956, pp. 1109 to 1117. 1 sheet of drawings I 209 617/210 6.62