DE1263836B - Device for obtaining interpolation values of digital signals available in the form of zero crossings of phase-modulated oscillations, which are fed to a main and auxiliary counter - Google Patents

Description

FEDERAL REPUBLIC OF GERMANY

GERMAN.

PATENT OFFICE

EDITORIAL

Int. Cl .:

H03k

German KL: 21 al - 36/22

Number: 1 263 836

File number: N 29559 VIII a / 21 al

Filing date: November 26, 1966

Open date: March 21, 1968

The invention relates to a device for obtaining interpolation values in the form of Zero crossings of phase-modulated oscillations available digital signals, with two phase-modulated signals have a mutual phase difference that is a whole A multiple of 180 ° and is preferably an odd multiple of 90 °, where the number the zero crossings is measured digitally in that these phase-modulated signals in the form of Pulses of a counting device are offered and the pulses as a result of the phase modulation of the Signals are counted positive or negative and form a sum, which is the number of zero crossings in discrete form, with interpolation between zero crossings being achieved.

Such a device is known from French patent 1 308 993, in which the phase modulated Signals from photoelectric detection elements in a distance measuring device with Grids. The number of zero crossings occurring in a certain direction is a discrete measure of the size of a distance in this direction. With such devices it is possible, as already mentioned in the earlier French patent 1 299 209, by interpolation to achieve a more accurate measurement result. The solution of the way as an interpolation in such and similar devices in which a signal of the type mentioned above is available, must be carried out, but has not yet been given.

With devices in which the information is available in discrete form, the problem arises how an easily and clearly interpretable measurement result can be achieved when an interpolation is used. For easy reading and greater accuracy of measurement results, it is desirable that a measurement value is made up of the discrete measurement value and an interpolation value, which is either subtracted from this discrete measurement value or added to it and the size of the interpolation value never greater than the distance between two consecutive discrete values and is preferably at most half of this distance. If z. For example, if the specified discrete value is 10 units, the distance between two discrete values is 5 units and the real value is 12, an interpolation must be made between the values 10 and 15, with the result being 10 units as a discrete measured value with + 2 units as Interpolation value or, if necessary, 15 units as a device for achieving
Interpolation values of in the form of
Phase modulated zero crossings
Vibrations available digital signals that a main and auxiliary counter
are fed

Applicant:

N. V. Philips' Gloeilampenfabrieken,

Eindhoven (Netherlands)

Representative:

Dr. H. Scholz, patent attorney,

2000 Hamburg 1, Mönckebergstr. 7th

Named as inventor:

Eric Tapley Ferguson,

Gerardus Cornells Maria Schoemaker,

Eindhoven (Netherlands)

Claimed priority:

Netherlands of November 30, 1965 (15 518)

discrete measured value with -3 units as interpolation value.

The counter for the discrete measured value steps must therefore follow the measurement result as closely as possible, see above that the interpolation always takes place between two successive discrete steps. In addition, if the discrete measured value jumps by one step due to a change in the measured value, the interpolation value jumps at the same time. If these conditions are always fulfilled, the uniqueness of the measurement result is also guaranteed.

The invention provides an apparatus in which these needs are met and it exists in that the counting device consists of a main counter and an auxiliary counter with limited counting capacity exists, with the impulses being offered to the auxiliary counter and at the highest or lowest final reading of the auxiliary counter, the further positive or negative counting pulses to the main counter are passed on, the main counter reading in discrete form the number of zero crossings represents and also connected to the auxiliary counter a device in which a

809 519/572

Voltage is generated, which is positive or negative, depending on whether the auxiliary counter has a value above or under his middle indicates the ratio of the successive time intervals while whose positive or negative voltage is present, a measure of the interpolation value between two successive main counter readings, and this ratio by means of a known per se Device is made visible.

In order to clarify the invention, the processing of the in a device is first of all, for example according to the French patent 1 308 993 achieved electrical signals explained again. It should be pointed out that the invention does not apply to this application example of the interpolation method limited.

In the device according to French patent specification 1 308 993, the coordinate χ is measured by means of two electrical signals S ₁ and S ₂ , which originate from photoelectric detection elements and have the following mathematical forms:

S ₁ = A cos - ^ - (- ^ - + c sin ü t

25th

S ₇ = A sin -i-

■ d

+ c sin ü t

35

To simplify the following discussion, the amplitudes of S ₁ and S ₂ are chosen to be the same, although this for the mode of operation of this method, which only depends on the zero crossings

Makes use is not necessary at all. - ^ - c represents the amplitude of the variable part of the phase of these signals, which changes sinusoidally with a circular frequency Ω means that there is no change in the distance, that there is no or only a small DC voltage component in the signal. Theoretically, it can be demonstrated from a series expansion of the signals that point M is counted back and forth. In this example drawn, for example. B. two steps there and two steps back in one direction along the arc and one step forward and one step back in the other direction. In a device according to the invention this back and forth counting takes place in an auxiliary counter which has a limited counting capacity of four levels, e.g. B. 1 and 0, and -1 and -2. The sum of the impulses in the subcarrier is therefore zero with a fixed χ.

If χ changes in a certain direction, then the argument of N, i. h / is the number of quadrants that the center M has traversed from a starting point where χ = 0 in this direction

Measure for -4- in the form of a discrete value, which in d

the device according to the invention appears on a main meter. This is achieved in that at If the value of χ changes in a certain direction, the center M begins to rotate in a direction corresponding to this direction of displacement, whereby the auxiliary counter always reaches the end of its counting capacity on the positive or negative side. The impulses not picked up by the auxiliary counter, so-called "overflow impulses", are activated immediately taken from the main counter.

So far, only the zero crossings of the signals S ₁ and S ₂ , that is, when S ₁ = 0 and S ₂ = 0, have been used to determine the distance χ in the known device. With each zero crossing, a counting pulse is counted positively or negatively, depending on the argument

40

= 0 must be selected, where J _{0 is} the

Bessel function of the zeroth order is: the first two roots are c = 1.531 and 3.514, where c = 1.5 and 3.5 are sufficiently good approximations in practice, d is a system quantity, namely a sixteenth part of that in the French U.S. Patent 1,308,993 mentioned raster period length p. If ρ = 10 μΐη, then d - 0.625 μΐη, and the dimension χ is measured in units of 0.625 μΐη.

The signals S ₁ and S ₂ represent a circular arc in combined form. In FIG. 1 this arc is shown for the case that c = 1.5, where M represents the center of the arc. In this case, the length of half the circular arc is due to the phase modulation

of the signal equal -

c, in this case equal to -j- π

or, in other words, the entire arc with M as its center occupies three quadrants. Each crossing of the axis means a zero crossing of one of the signals, and these zero crossings are counted in the form of pulses in a counting device. With a fixed value of χ, the phase modulation with an angular frequency Ω always increases or decreases. The position of M is therefore only determined within one quadrant, which means that in the measured value of χ there is an uncertainty of maximum

± - exists. However, the signals S ₁ and S ₂ contain

sufficient information for a more precise determination of the distance x. This can be done by interpolation.
The invention is based on the drawing

45. for example explained in more detail. This is done by a Device according to the. French patent 1 308 993 assumed phase-modulated signals as the source. In this context it should be pointed out again that the invention not limited to this device as a signal source.

Fig. Figure 2 shows a block diagram of a device according to the invention;
Fig. 3 illustrates the processing of the information of the argument of the mentioned electrical signals S ₁ and S ₂ .

In Fig. 2, 20 denotes the device according to French patent specification 1 308 993, in which the signals S ₁ and S _{2 are} generated. These signals are converted by a detector 21 known from the aforementioned patent specification in the pulse series corresponding to the zero crossing of these signals. The pulses obtained are counted positively or negatively in an auxiliary counter 22 with limited capacity according to the increase or decrease in the argument of the relevant signals S ₁ and S _2. At the highest or lowest reading of the auxiliary counter, further positive or negative counting

pulses passed to the main counter 23. According to the direction of displacement, the sum of the pulses on the main counter 23 will increase or decrease. The reading H on the main counter 23 shows how much the distance χ is a multiple of the dimension (/ or in other words, how many quadrants the center M of a circular arc with a length

from 2 - ^ - c in one direction or another

has gone through. It should be noted that this multiple of the dimension d, if d is, for example, 0.625; jm, can easily be converted into a simpler unit using a known conversion method. The signals S ₁ and S ₂ can, if desired, be made visible in composite form on a display device 24, with a fixed value of χ a as already in connection with FIG. 1 mentioned arc appears. With the auxiliary counter 22, the devices 25 and 26 are connected, in which the signals A and B to be discussed in more detail are obtained. The signals A and B are fed to a display device 27 and a control device 28, respectively. The control device 28 is in turn connected to the device 20. The operation of the device according to Fi g. 2 is based on Fi g. 3 described in more detail. In Fig. 3 is, on a vertical axis, the argument of the signals S ₁ and S ₂ in the form of a quantity

y = -j + c sin m plotted as a function of the angle α t shown on a horizontal axis. Here, for example, c = 3.5 is selected, and the auxiliary counter 22 in FIG. 2 can e.g. B. 3, 2, 1, 0 and

- Display 1, —2, —3 or —4, according to the

Position on the arc, which in this example is 1 j circle long. It should be noted that the circle radius caused by the amplitude A of the signals S ₁ and S _{2 is} not important, since only the zero crossings are important for the measurement. As a result, the interpolation z. B. regardless of the temperature of the photocells used, which is a main condition when using photo-sensitive semiconductor elements. In Fig. 3 is in FIG. 3a shows the graph of the signals S ₁ and S ₂ , which follows easily from the given formula. Sign change of S ₁ and S ₂ caused from each of the argument by changing, as g in F1. 3c, a count pulse can be derived positively or negatively. This pulse is recorded on the auxiliary counter 22 of FIG. The auxiliary counter is in the case of Fi g. 3c count between its outermost displays —4 and +3 with the same angular frequency .Q as the quantity y to and fro. On the vertical axis of the F1 g. 3 are a display H of the main counter 23 of FIG. 2 and the displays h of the auxiliary counter 22 plotted ^ If the distance χ is the whole number H times as large as the dimension d, the display H of the main counter is a direct dimension for x. y is always between H- 4 and H + 4. If y comes outside of this range, the "overflowing" counting pulses are passed on to the main counter 23, which jumps to H + 1 or H-1 according to the direction of displacement or continues if necessary. In this way, the main counter 23 is always set in such a way that y changes between the limits H + 4 and H-4, the auxiliary counter always between +3 and

- 4 counts.

If in this example c is exactly 3.5, the main counter 23 will display H for so long

between [H - yj and [H +

5
lies. So H is a quantized measure for the value

Ύ ' ^where ' ^{3ei x} measured ^with ^ ^he smallest unit of measurement d

The above only occurs when the amplitude c of the part of the argument that changes with the angular frequency Qt is always almost constant. In general, a constant c value is required for a good effect, even with other numerical values than those given above, for example. This will be discussed in more detail below.

The fact that the main counter 23 is always set in the example described in such a way that y changes between the limits H + 4 and H- 4, with the main counter counting between +3 and -4, is necessary to achieve an interpolation between the narrowest limits, in this case between H and H + 1 or H-1, essential.

An interpolation can take place as follows: If the value χ is not a whole multiple of the dimension d ,

z. B. -r = H + u, where H is an integer d

after the preceding is displayed on the main counter, and α is a number between the

^{B B} ' _l

Limits + - = - and - ~ - lies, then according to the

2 2 ^b

A signal A from the display of the auxiliary counter 22 is found in the device 25, the mean value of which is shown on a display device 27 as a measure for. ^ ^^ ^ ^ ^. ^ Signal A, in Fig. 3d eg symbolically represented by +1, derived from the auxiliary counter if the display of the auxiliary counter is 0, 1, 2 or 3, and a negative signal A represented by -1 , if the display of the auxiliary counter is -4, -3, -2 or -1. The device 25 is, for. B. a gate circuit known per se, which distinguishes these two display groups of the auxiliary counter 22. The device 27 is a known device for determining the mean value of a block signal. In this way ffi. ^ _{Value of χ or} χ _{ej becomes a unique value}

d
with a display H on the main counter 23 and a value of α read off the display instrument 27. Here, α can be positive or negative, according to the display H of the main counter 23 compared _{to the real value} * _{namely H} <^. It is

d d

namely possible that the display H, as already mentioned above, _{by a maximum of} _ 1 _or 1 _{from Wm} χ _afe _

2.2 d

can give way. This makes it possible to interpolate between the discrete values of the main counter 23, whereby the measured value by z. B. a factor of 10 can be determined more precisely.

As already mentioned above, the amplitude c must be particularly well constant for the device to work effectively. For this purpose, according to the idea of interpolation, a signal B can be taken from the auxiliary counter 22, which signal can be used as a control signal for the amplitude c. However, this is only possible for values of c equal to or greater than 2, since c is close to

1.5 there is insufficient information to take the mentioned signal B from the auxiliary counter. In order to meet the requirement of a constant amplitude in the case of c «1.5, further means, not specified in detail, can be used, with no use being made of the counting information of the auxiliary counter.

Another advantage of choosing c = 2 or higher for the amplitude of the phase modulation is the following.

In the example according to French patent specification 1 308 993, a critical speed occurs when the screen frequency becomes equal to the modulation frequency Ω due to a shift. In this case, a direct voltage component arises in the signal which is not allowed through by the usual alternating voltage amplifiers, so that the zero level is shifted and therefore zero-crossing discrimination is at risk.

The amplitude of this DC component

is at most J ₁ (γ c> ; a good choice of c are e.g. those values for which J ₀ (γ c j χ 1, such as c = 1.5 or 3.5. With c = 1, 5 or 3.5, this function J ₁ (-§- <) has the values 0.52 or -0.34. The risk of errors occurring is much smaller in the latter case.

For controlling the amplitude of the in Figure 3e. For example, in an apparatus 26 is symbolically drawn by +1 reproduced signal B the auxiliary counter when the display -4, -3, +2 or +3, while the auxiliary counter is a signal B = 0 is taken from the display —2, —1, 0 or +1. From the graph «it can be deduced that as c increases, the mean time value of B increases. This increase can then be processed in a known manner in a control device 28 to form a control signal which is fed to the device 20 for controlling the amplitude c. The change in the time average of signal B,

χ 1

if -; - H = α at c = 3.5 between - ^ - and a 2

+ - = - changes is negligible.

Claims (1)

Claims: 1. Device for obtaining interpolation values of phase- 45 modulated oscillations available digital signals, whereby two phase-modulated signals have a mutual phase difference which deviates by a whole multiple of 180 ° and is preferably an odd multiple of 90 °, the number of zero crossings being measured digitally by the fact that these phase-modulated Signals are offered in the form of pulses of a counting device and the pulses are counted positively or negatively as a result of the phase modulation of the signals and form a sum which indicates the number of zero crossings in discrete form, an interpolation between zero crossings being achieved, characterized in that the Counting device consists of a main counter (23) and an auxiliary counter (22) with limited counting capacity, the pulses being offered to the auxiliary counter (22) and the further positive or negative counting pulses for the auxiliary counter (22) at a certain positive or negative final reading Main number Counter (23) are passed on, the main counter reading in discrete form indicating the number of zero crossings, and a device (25) is also connected to the auxiliary counter, in which a voltage is generated that is positive or negative, depending on the auxiliary counter Indicates value above or below its mean, and the ratio of the successive time intervals during which a positive or negative voltage is present, a measure of the interpolation value between two successive main counter readings, and this ratio is visible by means of a known device (22) is made. 2. Apparatus according to claim 1, characterized in that in a device (26) accordingly the display of the auxiliary counter (22) a signal is generated by means of a device (28) as a control signal for the amplitude of the phase modulation of the phase modulated Signals. 3. Apparatus according to claim 2, wherein the maximum phase deflection of the phase-modulated signals has a value of -2- c , thereby characterized in that c is at least approximately 2, and preferably about 3.5. 1 sheet of drawings 809 519/572 3.68 © Bundesdruckerei Berlin