JPH0227601B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a digital angle coder.

An angle is generally determined by the relative angular position or displacement of two members about an axis of rotation. One member that determines the angle is fixed and used as a reference member. In this case the other member is rotatable and the angle relative to said reference member is measured.

In a method of digitally coding such an angle, a method has already been proposed in which an auxiliary member is used which rotates relative to the two angle determining members. The axis of this auxiliary rotating member (second rotating axis) is arranged as close as possible to the axis of the angle determining member (first rotating axis). The auxiliary rotary member has an index operable in association with each angle-determining member in one revolution relative to the two angle-determining members. In other words, for each rotation, the coincidence between the fixed angle determining member and the index and the coincidence between the movable angle determining member and the index are respectively 1.
You can get it one time at a time. Measuring the phase difference (time delay) of these two coincident signals provides information regarding the magnitude of the angle being coded.

For example, some devices use two current collecting brushes as the two angle determining members and a conductive area around one rotating cylinder as the auxiliary rotating member, and combine them to obtain a coincidence signal. Such a device is described in German Patent No. 595,939.

In conventional devices of this type, the relative speed of the auxiliary rotating member to the two angle determining members is kept almost constant, but since the auxiliary rotating member has only one index, the measurement accuracy is lower than that of the auxiliary rotating member. The disadvantage was that it was directly affected by the accuracy of the rotation speed.

The present invention solves these drawbacks, and provides a digital angle coding method and digital angle that make it possible to measure angles with high accuracy without being directly affected by fluctuations in the rotational speed of the auxiliary rotating member. It provides a coder.

In the present invention, the first indicators TF, TF0, TF180
and causing relative rotation between the first indicator group including the second indicators 6, TM, TM 180 and the second indicator group including at least the third indicators 5, 50. At this time, each indicator is a first indicator TF, TF0, TF0,
TF180 and second indicator 6, TM, TM
180, or a second group including at least one third indicator related to the second rotation axis, and further, the indicators of either group include: It is assumed that a series of reference marks 51 and 52 are formed at equal intervals around the first rotation axis. Then, when the reference mark of the indicator of the other group passes in front of the indicator of one group, matching of the two corresponding series is detected.

Then, by measuring the phase difference between the coincidence signal of one series and the adjacent coincidence signal of the other series, the first indicators TF, TF0, TF18
0 and the second indicator 6, TM, TM 180, the angle defined by the angular displacement formed around the first rotation axis 7 is coded into a digital value.

When measuring the phase difference, the total phase displacement obtained between the coincidence signal of one series and the adjacent coincidence signal of the other series is calculated between the indicators of the first group and the indicators of the second group. averaged over one or several complete revolutions between.

Therefore, for example, the reference mark (code) of each series
Errors due to interval variations can be eliminated. Also,
A typical error decreases in a certain relationship with the increase in the number of measurements, and since the present invention calculates the average value of a large number of measurement values as described above, it is possible to reduce the error due to chance. The influence of errors can be reduced. Therefore, even if there is an instantaneous variation in the rotational speed of the third indicator (auxiliary rotating member), there is an advantage that the third indicator (auxiliary rotating member) is not directly affected by the variation. This point is a significant difference from the prior art.

Reference mark in the present invention (hereinafter referred to as code)
can be provided on a rotating disk and read magnetically or optically. A device for reading codes consists of at least one reading head, for example a magnetic reading head or an optical reading head. Each code column is read by a corresponding read head.

In one embodiment of the invention, the rotary member as the third indicator has two or more rows of symbols provided on either or one side of its rotary disk. The angle-determining members as first and second indicators are each provided with a reading head, so that one head is fixed with the first angle-determining member and the other head rotates with the second angle-determining member. It has become freely possible.

Preferably, another movable head is provided symmetrically to the movable head with respect to the angular displacement axis (first axis). These movable heads read the same series of codes, and by calculating the average value, errors caused by misalignment between the angular displacement axis (first axis) and the relative rotation axis (second axis) can be corrected. Of course, it is also possible to arrange a large number of movable heads regularly around the axis of angular displacement.

A counter receives a clock signal of a predetermined frequency in between to determine the time displacement between one series of code reading signals and the other series' adjacent code reading signals. In this case, for example, the code reading signal is transformed into a square wave signal, and the time displacement is determined based on the time interval of two corresponding square wave signals.

As mentioned above, the digital information that is ultimately visually displayed in the angle coder is obtained through various methods including minute angle measurements performed using the average time displacement of two series of signals detected during one or several revolutions, and This method is a combination of rough angle measurement performed using the above method.

In an embodiment described in detail later using a magnetic reading method as an example, the coarse angle measuring device is
For example, it is composed of an up/down counter, and counts the difference in the number of two series of code reading signals between the angular displacements of the angle determining members (first and second). rotating member,
For example, a rotating disk is driven by a motor, the speed of which is related to the frequency of the counter clock. The rotating disk is connected to a synchronous motor controlled by the counterclock, for example.

Other embodiments of the invention are intended to be even more precise and simpler than those described above. In that example, two additional
Heads (auxiliary heads) are provided on the angle determining member, and further codes (auxiliary codes) are provided on the rotary member, so that each rotation produces one code reading signal (auxiliary code) different from the above-mentioned code reading signal. signal).

The angle sensing device comprises one or several flip-flop circuits, which are connected to the above-mentioned auxiliary head in order to change its state with each revolution. Then, due to the change in the state of this device, 1
Determine a time window corresponding to a revolution or several revolutions.
(This time window corresponds to the time for determining the average value.) The coarse angle measuring device measures the delay time between the code reading signals (auxiliary signals) from the two auxiliary heads for each rotation. Just use the .

Other features and advantages of the invention include:
It will become clear from the following detailed description with reference to the accompanying drawings, FIGS. 1-7.

A first diagram illustrating an embodiment of the angle coder of the present invention.
In the figure, a crystal clock circuit 1 with a frequency of 23.976 MHz divides the frequency of 23.976 MHz into 1/1000 and sends it out.

This frequency is sent to a frequency divider circuit 2 (for example a numerical counter), the output of which sends out a frequency of 74Hz.

This output frequency is further sent to circuit 3. This circuit is, for example, a power supply circuit for an eight-pole synchronous motor 4. This synchronous motor is available in the 9 904 110 09601 series or the 9 904 series depending on the direction of rotation.
110 09611 series polymotor type can be used. Motors of this type are commercially available from the company Société d'Etudes et de Development d'Apaire Electromechanique (SEDELEM). This motor with 8 pairs of electrodes has 7
It has a rotational speed of 4/8 = 9.25 revolutions/second, or 555 revolutions/minute.

This synchronous motor is connected directly to one of the discs 5 and rotates the disc at said speed. Disk 5 is a magnetic disk with an outer diameter of 18 mm. This magnetic disk is made by hollowing out a round aluminum plate and coating both sides with a magnetized film of iron oxide, Fe ₂ O ₃ , and is of the type used in the manufacture of magnetic disks for computers.

There is one magnetic head on one side of this disk 5.
T _F is installed. This head is of the write/read type. This magnetic head T _F
are placed on the disk 5 at such a position as to form a magnetic track with a median diameter of 13 mm. Therefore, the movement speed of the disk under the head is 380mm/
Seconds.

The magnetic head T _F floats on the flow caused by the rotation of the disk and is kept at a very close distance (several microns) from the disk. This fluid may be, for example, atmospheric flow or low viscosity oil. In order to place the magnetic head on the fluid, the dimensions of the head itself are extremely small, on the order of millimeters. In order to give the magnetic head buoyancy, wings are attached to it so that it floats above the magnetic disk 5 by itself. A floating magnetic head manufactured by Applied Magnetics, California, USA may be suitable.

Similar to the fixed magnetic head T _F , a pair of movable heads T _Mp and T _Ma are located on opposite sides of the disk. (For clarity, these two movable heads are shown far apart from the disk 5 in FIG. 6
The movable shaft 7 rotates around the angular displacement axis. It is desirable that this angular displacement axis be in line with the rotation axis of the disk 5 as much as possible. magnetic head
T _F and magnetic head T _Mp are connected to the magnetic mark recording circuit 8.
Connecting wires are attached to these heads to connect them. Recording circuit 8 receives a clock frequency of 23.976MHz.

It is better to manufacture this entire recording device in a factory. Commercially available angle coders do not include circuit 8. Therefore, in FIG. 1, this part is surrounded by a dotted line.

To perform this recording, first, the synchronous motor 4 is rotated at a speed of 9.25 revolutions/second determined by the crystal clock circuit 1. A recording circuit 8 of a conventional type is used to feed a signal with a frequency of 23976 Hz to the magnetic heads T _F and T _Mp . These signals may be square waveform signals. As will be discussed later, sinusoidal signals are used in high density recording devices.

The recording operation continues during one rotation of the disk. During this recording operation, magnetic heads T _F and T _Mp remain absolutely fixed relative to each other.

In this way, magnetic head T _F and magnetic head
T _Mp contacts each magnetic track on each side of the disk 5 with 2592 periods or marks over one complete rotation of the disk, each period corresponding to 500 seconds (in degrees) for a diameter of nearly 13 mm. .

It is better for the heads to stop in opposite positions. This results in identical and symmetrical inscribed magnetic tracks on both sides of the magnetic disk.

Here, the first indicator of magnetic head T _F and the second indicator of magnetic head T _Mp ,
The first group is constituted by the second auxiliary indicator of T _Ma . The second indicator and its auxiliary indicators are arranged symmetrically with respect to the angular displacement axis of the shaft 7.

The disks 5 constituting the second group have recording indicia or marks serving as third and fourth indicators on their magnetic tracks on both sides. In this regard, it would of course be possible to place the two heads on the same side of the disk and use the same magnetic track, provided that the movable head is not impeded by the fixed head during its movement.

Disk 5 (second group) is synchronous motor 4
The magnetic head (first group) is rotated by the magnetic head (first group).

Magnetic head T _F that reads two magnetic tracks
and T _Mp detects the average value of the displacement of the mark signal read during transition in the same direction as the recorded square waveform signal or movement in the same direction as the sinusoidal waveform signal, for example.

Therefore, each magnetic head reads a series of marks during one complete rotation of the disk 5 and repeats the reading. Auxiliary magnetic head
Considering that T _Ma reads the same magnetic track as the moving magnetic head T _Mp , by averaging the readings of the moving magnetic heads T _Mp and T _Ma , we can calculate the angular displacement axis of both heads. Non-linearity with respect to the rotation axis of the disk 5 can be corrected.

The three magnetic heads T _F , T _Mp and T _Ma are connected to a circuit 9 for reading and measuring the phase difference. A detailed explanation of the circuit 9 will now be given with reference to the accompanying drawings.

In FIG. 2, magnetic head T _F is connected to readout amplifier 11, and movable heads T _Mp and T _Ma are connected to respective readout amplifiers 12 and 13, which are identical to amplifier 11. In FIG. Each of these reading amplifiers first of all comprises a peak value limiter amplifier which transforms the sinusoidal signal into a square waveform signal. These square waveform signals are then differentiated and transformed into a waveform that results in a single pulse of a predetermined duration in the same direction, for example in the upward direction.

The pulses of the circuit 12T _Mp and the pulses of the circuit 11T _TF are applied to the anti-coincidence circuit 21. This circuit 21 serves to separate the two pulses arriving simultaneously at these two heads.

The two outputs T _F ′ and
T _Mp ' is applied to the two inputs of the addition/subtraction counter 22, which calculates the difference between the number of mark signals read by the movable head T _Mp and the number of mark signals read by the fixed head T _F , and also determines the direction of travel. or determine the sign of the angular displacement.

Circuit 23 controlled by phase measurement control circuit 15
receives the output of the clock circuit 1 along with the two outputs of the anti-coincidence circuit 21, and receives the mark signal or period read by the fixed head T _F and the mark signal or period read by the movable head T _Mp during one revolution of the disk 5. Measure the average phase difference of

The counter 22 and the phase difference measuring circuit 23 are connected, and the traveling direction information is mainly transmitted to the circuit 23.

The measuring circuit 23 counts clock pulses during the time interval between the pulse (mark signal) from the head _TF and the pulse (mark signal) from the movable head T _Mp .

Under the control of the origin adjustment circuit 16, the digital storage circuit 24 records the phase difference at the origin φ ₀ .

A digital subtracter 26 is connected to the measuring circuit 23 and the storage circuit 24, and the difference between the origin phase difference and the subsequent final measured phase difference is obtained here.

Furthermore, the circuit 27 passes some of the step outputs of the addition/subtraction counter 22, and adds or subtracts the counted value by one unit according to the sign of the phase difference.

In Figure 2, the auxiliary movable head T _Ma and the fixed head
A similar circuit is provided for the pulses output from T _F. Anti-coincidence circuit 31, addition/subtraction counter 32, average phase difference measurement circuit 33, origin phase difference storage circuit 34,
These are a phase difference subtraction circuit 36 and a unit correction circuit 37.

Also shown in FIG. 2 is a circuit 18 which averages the fine measurements coming from circuits 26 and 36 and the corrected coarse measurements coming from circuits 27 and 37. Although all these circuits output multiple digital information, circuit 18 clearly outputs only one binary average value.

This average value for the two moving heads is
9, the phase difference corresponding to a common angular displacement of the two movable heads relative to the fixed head is visually displayed.

In order to explain the coder of the present invention more clearly, the explanation will be made using a theodolite, assuming that the phase difference between the first aiming position and the second aiming position of the telescope is, for example, a vertical angle.

First of all, it is assumed that the magnetic heads T _F , T _Mp , and T _Ma are rigidly fixed relative to each other and that all heads read exactly the same number of cycles per second. Addition/subtraction counter 22 under such conditions
and 32 oscillate one unit for their initial values.

Regarding the first aiming position of the telescope of the theodolite, the phase control circuit 15 and the origin adjustment circuit 16 are activated. Circuit 16 sets addition/subtraction counters 22 and 23 to zero. The average phase difference between the pulses from each movable head and the fixed head is read (circuits 23 and 33 are activated, respectively).

Memory circuits 24 and 34, which are always affected by the origin adjustment circuit 16, record the origin phase difference obtained by the two movable heads in each case with respect to the first position of the telescope.

Circuits 15 and 16 are then deactivated;
Move the theodolite from the first aiming position to the second aiming position.
A relative movement occurs between the two movable heads and the fixed head. This relative movement creates a difference between the number of mark signals read by the movable head and the number of mark signals read by the fixed head. This difference for each movable head is calculated by adding/subtracting counters 22 and 3.
recorded in 2.

When the telescope reaches the second aiming position, only the phase measurement control circuit 15 is actuated anew, which causes a new average phase difference measurement in the measurement circuits 23 and 33. .

Thereafter, the two phase differences are produced by subtractors 26 and 36. In each correction circuit 27 and 37, a sign is assigned to these phase differences.

Circuit 18 averages the readings of the two movable heads and circuit 19 finally displays the numerical value of the angular displacement of the telescope from the first aiming position to the second aiming position.

As already mentioned, the mark signal or period obtained after counting by circuits 22 and 32 and correction by circuits 27 and 37 corresponds to 500 seconds (in degrees).

The phase difference measurement by the circuits 23 and 33 is carried out at 1/1000th division of the period, that is, at an angle of half a second. Therefore, the accuracy is obtained in seconds (angles).

Some particularly important circuits shown in FIG. 2 will now be described in more detail. First, let's start with the anti-coincidence circuit 21. The importance of this circuit is that pulses are the only basis for phase difference measurements in this embodiment, so the fineness and discrimination of these pulses is critical. That's the point. In the following explanation, the same circuit is used for the movable head T _Ma , so
The explanation will be limited to T _Mp , and the index o of the reference number T _Mp will be omitted.

The anti-coincidence circuit 21 shown in _{FIG .}
receives input.

One of the two pulses mentioned above arrives a little earlier than the other, and is referred to as the front pulse, and the latter as the rear pulse. Also, the pulses overlap for a little more than half of their duration, and both pulses have the same duration. Under these conditions, Figure 3,
The operation of this circuit will be explained with reference to FIGS. 3A and 3B.

The two inputs T _F and T _M of this circuit are first applied to a NAND gate 2110, whose output is low only during the intersection of the two input pulses (row C in Figures 3A and 3B). ).

At the outputs of AND gates 2111 and 2112, pulses T _F and T _M are obtained, but the common part of both pulses has been removed (line d in FIGS. 3A and 3B).

The outputs of the two AND gates 2111 and 2112 are connected to the respective monostable circuits 2113 and 21
14, but both circuits are connected to the first pulse T _F
and have the same delay T ₁ equal to the duration of T _M . These monostable circuits are designed such that their output lasts this delay time T ₁ from the rising edge of its input signal (Figure 3A, Figure 3).
(line e in Figure B).

The outputs of monostable circuits 2113 and 2114 are sent to AND gates 2115 and 2116, respectively. These gates additionally receive respective input signals T _F and T _M .

AND gates 2115 and 2116 yield the following results. The pulse that first arrives at the gate is not actually subjected to any change, but the pulses that arrive at the gate later have their non-common parts removed by this gate (line f).

The pulse coming from the NAND gate 2110, i.e. the common part of T _F and T _M , is inverted by the inverter 2120 and then extended by the monostable circuit ₂₁₁₇ by a delay time T
line). The non-common part of pulses T _F and T _M coming from gates 2111 and 2112 is NAND gate 2
118 and 2119. these
The NAND gate also simultaneously receives the output of monostable circuit 2117. The output of one of these two gates maintains the level "1" and indicates which one of the pulses T _F and T _M arrives first (line i).

AND gates 2121 and 2122 select the first arriving pulse from the outputs of gates 2115 and 2116 (row j).

After inversion by inverters 2123 and 2124, AND gates 2125 and 2126 select the pulse that arrived later of the two pulses (row k).

As illustrated in FIG.
On the path of pulses T _F and T _M until they reach 5 and 2126, delay elements 2140 and 214
1, 2142 and 2143 are inserted.
These delay elements are intended to maximize the performance of the decoupling circuit by preventing the generation of short duration interfering pulses.

Monostable circuit 2127 is NAND gate 2110
to receive the common part of the two input pulses. On the output side, the beginning of the negative wave is delayed by the T ₂ value. As a result, the only low-level part is the last small part of the common part (the third
Line l of Figure A).

The output of this monostable circuit 2127 is applied to AND gates 2128 and 2129. Both gates also input the outputs of AND gates 2121 and 2122. As a result, the first arriving pulse is OR
It is sent to gate 2151 and its output becomes T _F ' (or it is sent to gate 2152 and its output becomes T _M '). This first pulse has its final portion cut off by the action of monostable circuit 2127 with delay T ₂ (line m in FIG. 3A).

Later arriving pulses are sent from gates 2125 and 2126. Whether the two input pulses match almost exactly is determined by
Illustrated by AND gate 2130.

If they match, the inverter 2131
activates AND gate 2132 and the second pulse passes through gate 2133 or gate 2134 depending on T _F or T _M . These two gates also input the output of monostable circuit 2135.

Single direction circuit 2135 transforms the inverted common portion formed by gate 2110 into a slightly shorter common portion with a constant duration longer than the delay time T ₂ described above (line l of FIG. 3B).

As a result, later arriving pulses have a delay T ₂
(as can be seen by comparing rows m of FIGS. 3A and 3B) until the earlier arriving pulse formed by monostable circuit 2127 has finished.

If the two input pulses are completely separated, there will be no common part and such pulses will appear at the outputs of AND gates 2121 and 2122, and AND gates 2125 and 2126 will remain at a low level.

When monostable circuit 2127 (time T ₂ ) ceases operation, gates 2128 and 2129 pass pulses in turn.

Finally, if the two pulses almost match, gates 2121 and 2122 will output low levels, and pulses T _F and T _M will pass through AND gate 2
125 and 2126, AND gate 2
130 continues to operate for the duration of these two synchronization pulses.

In that case, AND gate 2137 outputs pulse T _F at the beginning. This gate 2137 is the duration of the common part coming from the monostable circuit 2127.
Only the starting part of T ₂ is sent to the OR gate 2151, and the output T _F ' of this gate is output.

Similarly, AND gate 2134 is a monostable circuit T ₃
It acts on the output of 2135 and generates a pulse after a predetermined time.
This output is applied to OR _gate 2152.

Therefore, three types of operations are performed.

1. If two input pulses are completely separated by a certain amount of time, the anti-matching pulse will pass them through unchanged.

2 The two input pulses do not match perfectly, but
Monostable circuit if it has a common part
Separate them by T ₂ and T ₃ .

3. If the input pulses are almost perfect or perfectly matched, the pulse T _F ' is output first, followed by the short, completely separated pulse T _M '.

In FIG. 4, an embodiment of the addition/subtraction counter 22 of FIG. 2 is illustrated. This circuit also detects the sign of the rotational direction, that is, the angular position.

The circuit of FIG. 4 receives pulses T _F ' and T _M ' coming from the anti-match circuit of FIG. Bistable trigger circuit 2210 sends out heading information SENS as described later.

According to the state of the output terminals Q and of the trigger circuit 2210, the inputs T _F ′ and T _M ′ are
OR gate 2 via gates 2211 to 2214
221 and 2222 directly or inverted.

The outputs of OR gates 2221 and 2222 are applied to a series of four integrated circuits 2231-2234 via logic circuits 2241-2243. This integrated circuit is a binary coded decimal BCD addition/subtraction counting integrated circuit. The addition input terminal UP, subtraction input terminal DOWN and carry output terminal CARRY of these integrated circuits,
The borrowed output terminals BORROW can be connected in series using a known method. In Figure 4, this connection is connected to an integrated circuit sold by Texas Instruments.
This was done using SN74192.

To simplify FIG. 4, the four stage outputs of each integrated circuit, which are routed downstream in FIGS. 2 and 5, have been omitted in this FIG.

These integrated circuits are bistable trigger circuits 221
Similarly to 0, it is set to zero by the signal SCDM, that is, the measurement start adjustment signal sent out during the operation of the origin adjustment circuit 16 (FIG. 2).

The OR gate 2221 has an auxiliary input terminal SUCD, and inputs the unit subtraction signal for the addition/subtraction counter sent from the unit correction circuit 27 of FIGS. 2 and 5. Pulse SUCD appearing at OR gate 2221
is sent to the input terminal DOWN of the first counter 2231, and one unit is subtracted from its content.

The output zero of all the outputs of the addition/subtraction counters 2231 to 2234 is detected by the OR gate 2244,
This gate is the zero detection signal of the addition/subtraction counter 22.
Send out DZCD. This signal DZCD becomes a logical zero when all output terminals of the adder/subtractor 22 are set to zero. When the first pulse that arrives after the counter 22 has been set to zero is a subtraction pulse, perturb this trigger circuit by applying one pulse to the input terminal CLOCK of the trigger circuit D2210. Can be done. Inverter 2245 and AND gate 2246
creates such a situation.

The counting pulses sent from OR gate 2222 are normally reversed in NOR gate 2241 and counted at input terminal UP of circuit 2231. In the case described above, where the signal DZCD is zero and the first pulse that appears first is in principle a subtraction pulse, this pulse is actually an AND gate 2246 and a NOR gate operated by an inverter 2245. It is applied to the addition input terminal UP of the circuit 2231 via 2241.

The subtraction pulse or unit subtraction pulse SUCD transmitted from the OR gate 2221 is transmitted from the counter 2
Activated by 2DZCD zero set
AND gate 2242 and further inverter 224
The input terminal of the first counter 2231 via 3
Transmitted to DOWN.

In summary, trigger circuit 2210
Input information SENS of the displacement direction of T _M and fixed head T _F , that is, the angle sign. This circuit will oscillate if the first pulse is a subtraction pulse and the counter is set to zero. At that time, the pulses T _F ′ and T _M ′ are input terminals of the counter 2231.
Transmitted towards UP or DOWN.

The initial sign of the contents of the addition/subtraction counter 22 can be determined arbitrarily. Alternatively, it is determined by moving the movable head T _Mp in front of the fixed head into one subdivision in any direction.

If head _TM is displaced in the direction of disk rotation with respect to head _TF , the adder/subtractor will input +1 for each positive pulse read by _TF and -1 for each pulse read by _TM , so it will be zero. Move between and +1. In that case, the phase measurement is +
(positive) sign, and measurement starts from T _F.

If the displacement direction is reversed, the addition/subtraction counter will be zero and -
Move between 1. In that case, the phase measurement has −
(negative) sign is given and measurement starts from T _M.

After a negative displacement, the final phase measurement is made with respect to the fixed magnetic head T _F , depending on whether the count of the adder/subtractor ultimately becomes positive or negative.
Either it is done with respect to the movable magnetic head _TM .

In FIG. 5, the circuit 23 is shown above the broken line, and the already explained circuit 22, origin phase storage circuit 24, subtracter 26, and correction circuit 27 are shown together below the broken line.

In circuit 2315 a signal DICD is output, which changes state at the end of each pulse UP or DOWN as applied to the first circuit 2231 of the adder/subtractor. For example, at the end of pulse UP, signal DICD begins to change to level 1. At that time,
A counting of 2592 periods is performed with the start of this signal DICD, counting all transitions in the same direction. For this purpose, circuit 2315 includes a 2592 module counter. When this counter finishes counting 2592, it indicates that the phase difference measurement is finished.

Although the period counting circuit 2315 is provided with one output line SDD2, this output line is designed not to generate a phase difference waveform on the output side of the circuit 2311 when 2592 periods are not counted.

The start of period counting is performed by the phase measurement start signal VDMP coming from the phase measurement control circuit 15 of FIG. In response to this signal, circuit 231
5 is a control signal that sets accumulator 2314 to zero.
Generates SDD3. This accumulator calculates the average phase difference φM between the addition pulse and the subtraction pulse.

Finally, this counting circuit 2315 receives the above-mentioned measurement start adjustment signal coming from the origin adjustment circuit 16 in FIG.
Enter SCDM. This circuit records this signal SCDM at least until the end of counting and then
Output signal SDD1 is generated after counting of 2592 cycles (disc rotations) is completed.

While signals VDMP and SDD2 are present,
Circuit 2311 produces a phase difference waveform. For this,
Three signals CDA1, CDA2 and CDSA are generated separately using signals UP, DOWN and DICD.

The signal DICD pulses UP and DOWN (i.e.
T _F ′ and T _M ′) always arrive alternately, i.e. one arrives when the signal DICD is in one state and the other when the signal is in the other state. It's for viewing.

Pulse UP since the start of phase measurement (signal VDMP)
Assuming that pulses DOWN and DOWN arrive alternately, circuit 2311 generates one waveform 1 for each time interval from the end of pulse UP to the end of the immediately following pulse DOWN. signal of
Generates CDA1. This signal CDA1 is in this case the same as the signal DICD.

Furthermore, the signal CDA1 is represented by a time measurement value with a + (positive) sign and a coefficient of 1.

When pulses UP and DOWN no longer arrive alternately, a different state occurs.

When two pulses DOWN appear one after another,
One signal CDS containing one waveform from the end of pulse DOWN to the end of subsequent pulse UP
1 is only generated. and two pulses
UP appears one after another, and here is the first pulse.
A shift between DOWN and UP takes place.

The signal CDS1 is represented by a time measurement with a - (negative) sign and a coefficient of 1.

Finally, after a normal first pulse alternation,
Even if two pulses UP appear one after another,
A signal CDA1 is generated, but it appears in reversed form. That is, two consecutive pulses UP
Waveform CDA1 is inserted between the pulse DOWN and the pulse UP that appear after that.
Then two consecutive pulses DOWN appear,
This will set up the first police box. In that case, signal CDA1 is the same as signal DICD, even though the alternation is interrupted.

Furthermore, a signal CDA2 is also generated, which corresponds to a time measurement value with a + sign and a factor of 2.

This signal CDA2 contains one waveform during each time interval from the pulse UP to the immediately following pulse DOWN, ie, during the apparently abnormal time.

By combining the signals CDA1 and CDA2, a measured value corresponding to the actual value is obtained, despite their anomalies. In that case, the overlapping portions of the waveforms are calculated twice.

The logic gate 2312 is the phase difference waveform generation circuit 2
Divider 2592 for signals CDS1 and CDA2 coming from 311
is transmitted to the divider 2313. This logic gate 2
312 receives a clock frequency signal of 23..976MHz,
Divider 2313 divides this frequency by 2592 while waveform CDA1 appears, and then divides waveform CDA1 by 2592.
While 2 appears, its frequency is 1296 (2592÷
2), thereby generating the first output signal SDA2592 having a + sign. Divider 231
3 further divides the clock frequency by 2592 while waveform CDS1 is appearing to generate a second output signal SDS2592 having one sign. These output signals SDS 2529 and SDA 2592 are sent to the addition and subtraction input terminals of the average phase difference φM counter 2314.

Origin phase memory 24, which was initially set to zero by signal SCDM, records the contents of counter 2314 in response to signal SDD1. At that time, the origin phase is stored in the memory circuit 24, but this origin phase is further stored in the binary decimal system BCD.
is applied to a subtracter 26 which operates at .

As subsequent operations of circuit 15 occur, a new signal VDMP is generated and period counter 2315
operates again and activates the phase difference waveform generation circuit 2311. After 2592 cycles, the counter 2314
Calculate the final average phase difference, and this average phase difference is
It is applied to another input terminal of the BCD subtracter 26.

The BCD subtractor 26 outputs the average phase difference between the final position and the initial position.

Since the clock frequency is 1000 times the mark recording frequency on the disk, the phase difference measurement is performed with an accuracy of 1/1000. Therefore, subtractor 26 performs a BCD calculation with four decimal numbers.

This BCD subtractor operates, for example, by adding the complement to 9, taking into account units that are not at the least significant bit level. However, if the direction SENS changes between the stored origin phase and the measured phase φM, simply the two
Just add the values.

A unit correction circuit 27 for correcting the contents of the addition/subtraction counter 22 receives the carry and borrow outputs CARRY and BORROW of the average phase counter 2314, determines the sign of the average phase, and performs correction according to the sign.

Circuit 27 also receives one signal from BCD subtractor 26. This is because the addition/subtraction counter 22 is zero (DZCD signal) and the adjustment signal at the start of measurement.
It is also clear from inputting SCDM.

The correction circuit 27 forms a unit subtraction signal SUCD for the addition/subtraction counter from these various signals, and directly outputs the unit subtraction signal SUCD to the subtraction input terminal of the first stage 2231 of the addition/subtraction counter 22.
Apply to DOWN.

Once this correction is complete, circuit 18 of FIG. 2 averages the measurements of movable head T _Mp and movable head T _Ma . This circuit 18 performs the division by two in binary decimal format in a known manner.

Finally, circuit 19 displays the final phase difference in seven signed decimal digits.

When applied to a meridian, two angle coders of the present invention are naturally used, one for vertical angles and one for horizontal angles. Naturally, the electronic circuits have common parts, mainly the clock circuit 1, the origin adjustment device 16, and the control circuit 16.
(Figure 2) is included. In implementation, circuits 15 and 16 may be push-button operated, operated using direct traverse angle brakes, or both.

In another embodiment of the invention, a DC motor is used instead of a synchronous motor controlled by a frequency source to maintain the temporal regularity of the mark signal, e.g. to ensure regular reading of the mark by a fixed head. The electric motor is controlled in this way. When measuring the regularity of the spacing of marks on a recording disk, the rotational speed of the disk is virtually constant.

Of course, the magnetic detection method of the mark on the rotating disk,
It's not absolutely necessary. In addition, optical detection methods have high measurement accuracy, such as magnetic head T _F
It is also possible to use photoelectric elements instead of T _Mp and T _Ma to draw black and white lines on the rotating disk.

In fact, the preferred embodiment, which will be described below with reference to FIGS. 6 and 7, employs an optical reading system.

As is clear from FIG. 6, the group in which each indicator has a series of scale marks is a group that has nothing to do with angle measurement, ie, the second group described below.

In FIG. 6, reference numeral 40 designates a bearing which is part of the device, which is shown very simply within the frame of dashed line 41. In FIG.

The device 41, enclosed in broken lines, comprises a first indicator comprising two main reading heads T _Fp and T _F 180 for fine measurements and an auxiliary reading head T ₁ for coarse measurements.

The second indicator is connected to one rotating shaft 42 supported by a bearing 40. Rotation of the shaft 42 with respect to a bearing 40 connected to the device 41 provides the angle to be measured.

The shaft 42 has two main reading heads T _Mp for fine measurements.
and T _M 180, and auxiliary reader for rough measurements
It is integral with one radial element supporting T ₂ .

One disk 50 that absorbs light and therefore reflects little or no light is driven by an electric motor 55.
, and rotates about a single axis that substantially coincides with the axis of rotation of shaft 42 . If the two axes are not matched as precisely as possible, a residual deviation will occur, and the first rotating shaft that provides angular displacement to the shaft 42 and the second rotating shaft that drives and rotates the disk 50 will become distinguishable. .

The disc 50 has two primary indicators, which are two series of engraved reflective marks regularly distributed around the axis of rotation of the disc. These two series of marks are the two tracks 51
and 52. It is preferable that these two track marks are carved on the same radius on the disk, but it is not necessary to do so.

In some embodiments, the number of these marks may be
There are 2000 pieces, and the diameter of the disk is about 6 cm. The line engraving techniques used in precision optics today can produce 2000 lines per revolution on a circle with a diameter of about 5 cm or a circumference of 15.5 cm. This corresponds to a density of about 13 lines per millimeter. The thickness of the scored lines should preferably be 1/2 to 1/10 of the line spacing, for example, 1/4 of the line spacing.
But that's fine.

The disc is further provided with two auxiliary marks 53 and 54 on separate tracks, one per revolution.
mark. Further, although it is preferable that the marks be provided on the same radius, it is not necessary to do so.

The disk is driven and rotated at a speed of 2.5 revolutions/second. The frequency corresponding to one mark per revolution on tracks 53 and 54 is 2.5 Hz, and for the 2000 marks regularly distributed on tracks 51 and 52,
It is 2000 times the frequency of No. 4, or 5000Hz.

The reading head T _Fp includes one light emitting element 420, one grating 421, and one light receiving element 422. In FIG. 6, elements 420, 4 are designated by reference number _T
One axis line is drawn up to 21,422. This axis extends to the lower end area of mark track 51, which shows the relationship of these three elements.

The reading head T _F 180 is of the same construction, but its axis is located in a different area of the track 51.
That is, the area exactly opposite to that for the reading head T _Fp has been reached.

Furthermore, an auxiliary reading head T ₁ is installed next to the main reading head T _Fp , which is connected to the track 5.
This is for reading the mark 3. The auxiliary reading head T ₁ is of similar construction to the reading head T _Fp , but its grating 425 corresponds to only one thin transparent interval or slit. To standardize manufacturing, an identical slit 426 is provided near the grating 423 attached to the read head T _F 180. This auxiliary grid 426 is used, for example, when a second auxiliary reading head is installed as needed or as an auxiliary means.

Similarly, the second indicator also marks and tracks 5.
2 and a main reading head T _Mp .
An auxiliary reading head T ₂ is mounted next to this reading head T _Mp and is associated with the mark 54 . The reading head with respect to the axis of rotation of shaft 42
Another main reading head is located directly opposite T _Mp .
T _M 180 is installed and engages a portion of track 52. This track section is at the exact opposite position of the same track to the track section associated with the movable reading head T _Mp .

In this way, due to the structure, the movable heads T _Mp and T _M 180
are mounted symmetrically with respect to the axis of rotation of the shaft 42 which is subjected to angular displacement. Fixed head T _F 180
This fixed head is adjustable in position, as indicated by double-headed arrow 427 near support 428. This adjustment takes place in such a way that the fixed head T _F 180 is itself symmetrical with respect to the second rotational axis relative to the fixed head T _Fp according to its construction (in the case described below). Experiments have shown that by using two fixed heads for fine measurement, T _Fp and T _F 180, and attaching them in opposite positions in the same way as movable heads for fine measurement T _Mp and T _M 180, It was found that the accuracy was considerably high.

Of course, FIG. 6 is a pre-assembly drawing with parts shown separated for clarity.
In reality, both the grid of fixed heads such as 421, 423, 425 and the grid of movable heads such as 431, 433, 435 are provided in close proximity to the mark series on disk 50. The very complex structure of the mechanical part of a digital angle coder will become very clear from such a diagram.

The electronic part of the coder will now be described in detail with reference to FIG.

First of all, it goes without saying that the input side of this electronics is connected to the output side of the reading head.

As a light emitting element, for example, a photoelectric diode commercially available from Texas Instruments Inc.
Photodetectors that can use TIL23 or 24 include:
The optical transistor LS600, which is also commercially available from Texas Instruments, or the RTC
Optical transistor marketed in France by La Radiotechnique Complex
BP×71-C type can be used.

These elements are powered in a known manner and equipped with suitable optical systems. This optical system may be a manufacturer's product.

The circuits 61, 62, 71, 72, 81, 82 are therefore connected to the light receiving element.

Grids such as 425 and 435, as already mentioned, only correspond to one slit. Tracks 53 and 54 have at least one
The mark is read by passing through the slit. As a result, one pulse appears from circuits 61 and 62 of FIG. 7 for each revolution. These circuits are connected to the auxiliary fixed read head _T1 and the auxiliary movable read head _T2.
is connected to.

The gratings 421, 423, 431, 433 have the equivalent of 20 line marks, for example similar to the continuous marks of tracks 51 and 52.

When the mark series perfectly coincides with the grating, most of the light rays emitted from the light emitting elements will pass through the grating and be absorbed or transmitted by the disk 50.
Only a small portion of the light beam is reflected and enters the photodetector. This is because the marks on the disk and the marks on the grid matched. As the reflective mark array on the disk gradually deviates from the grating mark due to relative rotation, the amount of reflected light received by the light receiving element gradually increases until it hardly changes. Each mark on the track then falls completely between two marks on the grid. in this way,
When the mark series on the track matches the mark on the grid, the match is detected based on the minimum amount of light received by the light receiving element.

Although in principle only one mark is sufficient, the use of gratings such as 421 and 423 with 20 marks is intended to increase the signal-to-noise ratio at the output side of the light-emitting element, and at the same time This is to completely correct the unevenness of mark positions and achieve a kind of balance.

As already mentioned, the signals obtained by the micromeasuring head are compared to the fixed heads T _Fp and T _F 180
is output _in the form of an electrical signal to the outputs of circuits 71 and ₇₂ in FIG.

In FIG. 7, a rotational speed detection circuit 70 used to determine one rotation with respect to relative rotation.
The output of the reading circuit 61 for the fixed head _T1 is used in this. This circuit is equipped with one or more bistable trigger circuits and has a fixed head.
Changes its state in response to a signal from _T1 .
Between the first signal and the second signal, circuit 70 forms a "first rotation" output state on its output line that connects to AND gates 710 and 810.
710 and 810 also receive the outputs of reading circuits 71T _Fp fixed head and 81T _Mp movable head.

Between the second and third signals read by fixed head _T1 , circuit 70 forms a "second rotation" output state, which is connected to AND gates 720 and 8.
20. These gates are also connected to the outputs of reading circuit 72 (fixed head T _F 180) and reading circuit 82 (movable head T _M 180).

The outputs of AND gates 710 and 720 are combined in OR gate 73 and the outputs of AND gates 810 and 82 are combined in OR gate 73.
The zero outputs are combined in OR gate 83.

Circuit 70 thus provides a type of multiplexing, with heads T _Fp and T _Mp being used during the first rotation and heads T _F 180 and T _M 180 being used during the second rotation. This multiplexing operation is limited to only two revolutions. As will be described later, the phase differences between the signals are averaged during these two rotations. This operation is repeated several times to obtain several average measurements in succession.

Here, if we take the output of OR gate 73 as an example,
Sometimes this output is the signal from read head T _Fp and sometimes it is the signal from read head T _F 180. Moreover, it is clear that it consists of several signals that faithfully represent the rotation of disk 50.

These signals control a clock circuit 74 (VCO type or voltage controlled oscillator) with a phase control port. As described above, the clock circuit used to measure the phase difference between signals is not highly accurate, but is controlled in accordance with the rotation of the disk 50. The rotation of the disk 50 is, of course, maintained at a constant speed by, for example, a DC motor, but "uneven rotation" that occurs in the rotational movement of the disk 50 is detected by the clock circuit 74, which is It provides a so-called special time determined with respect to the rotation of . Such measures go a long way in increasing accuracy.

Returning now to the circuits 61 and 62 connected to the movable heads _T1 and _T2 , the outputs of these circuits are both applied to a single circuit 63;
A coarse measurement waveform is formed by this circuit. Circuit 63 therefore consists, for example, of a bistable trigger circuit, which is set into one state by the output of circuit 61 and into the other state by the output of circuit 62.

One of the status outputs of this trigger circuit is connected to circuit 63.
One coarse measurement waveform is output as the output. This waveform is fed into AND gate 64. gate 64
receives the output of the angle measurement start circuit 65 along with the 10 KHz frequency output from the clock circuit 74.

The output of circuit 63 is also transmitted to AND gate 66. This gate is 100K from clock circuit 74.
It also receives the output of the starting circuit 67 as well as the frequency of Hz. Only one of gates 64 and 66 operates.

During fine angle measurement operation, circuit 65 activates gate 64 whose output passes through OR gate 69 to counter 90 which takes a coarse measurement. Indeed, during the duration of the coarse measurement waveform formed by circuit 63, counter 90 receives 10K from AND gate 64.
Receive Hz pulses (controlled). By counting these pulses a coarse measurement is determined.

On the other hand, circuit 75 receives the outputs of OR gates 73 and 83 and forms a fine measurement waveform. These waveforms are the signals read by one of the fixed heads T _Fp and T _F 180 and the movable heads T _Mp and T _M
180.

These waveforms can be output from the output side (++) of the circuit 75 by selecting a sign set in advance.
Appears as (+) or (-). The selection of this sign determines whether the signal from the fixed head (first indicator) occurs before the signal from the movable head (second indicator) or vice versa. Of course, the selection of the code is free, but it is determined by the structure. Output (+
+) indicates that the waveforms overlap (ambiguity).

Output of circuit 75 (++), (+), and (-)
are the respective AND gates 80, 76 and 77
is applied to. Each of these gates also receives the same 10 MHz output from clock circuit 74.

The output of gates 80, 76 and 77 is the divisor 4000
is applied to counters 78 and 79 with dividers. The output of gate 80 is applied to the second bistable input terminal (input x 2) of counter 76. Since there are 2000 marks per revolution, these counters measure the phase difference of adjacent signals between heads T _Fp and T _Mp in the first revolution and between heads T _F 180 and T _M 180 in the second revolution. The average of

The outputs of dividers 78 and 79 are added to addition/subtraction counter 91
applied to the addition and subtraction inputs of . circuit 7
Under two rotations detected by 0, the counts of circuit 91 determine the average micromeasurement value, as already mentioned.

circuit 75 and gates 80, 76 and 77,
Dividers 78 and 79 include circuit 2311 and gate circuit 231 already described with respect to FIG.
2. It is similar to the addition/subtraction circuit 2313. The inputs coming from gates 73 and 83 are connected to circuit 2311
corresponds to the inputs UP and DOWN. Counter 78
and the output of 79 is the output signal of circuit 2313
Corresponds to SDA2592 and SDS2592.
Therefore, even though the anti-coincidence circuit is less complex than those described above and the absolute reference is determined by an auxiliary indicator, "ambigity cancellation" still occurs.

Also, similar to the above, a correction circuit 92 is inserted between the fine measurement circuit 91 and the coarse measurement counter 90, and when an error in the coarse measurement is revealed by the fine measurement calculation, the coarse measurement counter is Correct only the units.

Finally, the outputs of counters 90 and 91 are decoded and
It is sent to the numerical display circuit 95. The decoding operation continues, but the display is multiplexed by AND gate 94. This gate 94 receives a frequency of 1 KHz from the clock circuit 74 and is controlled by the display start circuit 93. This circuit 93 can also be operated by the angle measurement starting circuit 65 or the angular displacement starting circuit 67.

The operations described above with reference to FIG. 7 relate to precision angle measurements performed during two revolutions. The circuit 65 controls the initiation of the necessary operations and at the same time sets the various states necessary, in particular the rotation detection circuit and the counters 78, 79, 90, 91 to zero. After two revolutions, the angle measurements are displayed by circuit 95. The electronic circuit then waits for the next manual angle measurement control or repeats the same operation several times under the control of circuit 65, for example.

It has other useful uses besides angle measurement. It is a precision direction measurement. In this case circuits 66, 67 and 68
You can use . Since fine measurements take some time, it is difficult to continuously perform fine measurements many times until the desired accurate orientation is reached. Therefore, a "semi-coarse measurement" is performed using circuits 66, 67, and 68, and this is referred to as a coarse azimuth measurement.
In particular, this semi-coarse measurement is carried out without interruption. Since the starting circuit 67 stops the rotation detection circuit 70 when in the "first rotation" state, the entire micro-measuring circuit is in a stopped state. Clock circuit 74 is then controlled by fixed head T _Fp . While the rough measurement waveform 63 appears, the frequency of 100 KHz output from the clock circuit 74 is outputted by the counter 6 of the module 10.
8 and its output is sent through an OR gate 69 to a coarse measurement counter 90. In this case the clock frequency is 100KHz, so the measurements thus obtained will be more accurate. Finally, the azimuth measurement starting circuit 67 acts on the display starting circuit 93 to cause the circuit 95 to display the rough measurement value. This continuously repeated rapid (400m sec)
The operation is more advantageous if one obtains as close as possible to the desired exact heading before taking one or more complete fine measurements (each taking approximately 1 second) to obtain the exact heading. becomes.

The angle coder explained with reference to Figures 1 to 5 is an absolute type when making ambiguous fine measurements, and it measures the relative displacement of the angle determining member when making coarse measurements. It is a relative type because it is based on The angle coder of the embodiment described with reference to FIGS. 6 and 7 is a true absolute angle coder. This type of coder can detect angles without ambiguity at any position and provides an effective setting of 0.01 centimeter grade.

Of course, it is also possible to change this absolute type coding to relative type coding. In that case, an initial value is introduced by means of a coding wheel onto the parallel inputs of the precise measuring counter 90 and the parallel inputs of the fine measuring counter 91.

The coder according to such an embodiment is mainly applied to sutra rituals. As already pointed out with respect to the coders of FIGS. 1-5.

[Brief explanation of drawings]

1 is a schematic diagram of an angle coder according to an embodiment of the present invention, FIG. 2 is a main electrical system diagram of the electronic part of the angle coder of FIG. 1, and FIG. 3 is a detailed electrical diagram of the anti-matching circuit 21 of FIG. 2. System diagram, Figures 3A and 3B are time diagrams related to this circuit, Figure 4 is a detailed electrical system diagram of the differential period counting circuit 22 of Figure 2, and Figure 5 is the average phase difference measuring circuit of Figure 2. 23. Partial detailed electrical system diagram of the counting circuit 22, origin phase storage circuit 24, phase difference measuring circuit 26, and unit correction circuit 27, FIG. 6 is a mechanical parts diagram of a modified embodiment of the angle coder, and FIG. FIG. 3 is an electronic circuit diagram for performing angle measurement of the coder shown in the figure. Figure 1: 1... Crystal clock circuit, 2... Frequency divider circuit, 3... Power supply circuit, 4... Synchronous motor, 5... Magnetic disk, 6... Arm, 7...
Movable axis, 8...Magnetic mark recording circuit, 9...Mark reading/phase difference measuring circuit, _TF ...Fixed magnetic head, _TMp , _TMa ...Movable magnetic head. Figure 2:1
1... fixed head read amplifier, 12... movable head read amplifier, 13... movable head read amplifier,
15... Phase measurement control circuit, 16... Origin adjustment circuit, 18... Averaging circuit, 19... Phase difference display circuit, 21... Anti-coincidence circuit, 22... Addition/subtraction counter,
23...Average phase difference measuring circuit, 24...Digital storage circuit, 26...Digital subtracter, 27...
... Unit correction circuit, 31 ... Anti-match circuit, 32 ...
Addition/subtraction counter, 33...Average phase difference measuring circuit, 34
... Origin phase difference memory circuit, 36 ... Digital subtracter, 37 ... Unit correction circuit. Figure 3: 2110
...NAND gate, 2111...AND gate,
2112...AND gate, 2113, 2114
...monostable circuit, 2115, 2116...AND
Gate, 2117...monostable circuit, 2118,2
119...NAND gate, 2120...Inverter, 2121, 2122...AND gate, 2
123,2124...Inverter, 2125,2
126...AND gate, 2127...monostable circuit, 2128, 2129...AND gate, 21
30...AND gate, 2131...Inverter,
2132...AND gate, 2133, 2134
...Gate, 2135 ... Monostable circuit, 2137
...AND gate, 2140, 2141, 214
2,2143...delay element, 2151,2152
...OR gate. Figure 4: 2210...Bistable trigger circuit, 2211-2214...AND gate,
2221, 2222...OR gate, 2231~
2234... Integrated circuit for addition/subtraction counting, 2241...
NOR gate, 2242...AND gate, 224
3...Inverter, 2240...OR gate, 2
245...Inverter, 2246...AND gate, UP...Addition input (terminal), DOWN...Subtraction input (terminal), CARRY...Carry output (terminal),
BORROW...Borrow output (terminal), CLEAR...Erase input (terminal). Figure 5: 2311...Phase difference waveform forming circuit, 2312...Logic gate, 2313
...Divider, 2314...Counter. Figure 6: 40
... Bearing, 41 ... Device within the frame of the interrupted line, 4
2... Rotating shaft, 50... Disc, 51, 52...
...Reflective mark/track, 53, 54...Reflective mark/track, 55...Electric motor, T _F , T _F 18
0... Fixed head reader, TMo, T _M 180...
Movable reading head, T ₁ , T ₂ ... Auxiliary reading head, 420, 429 ... Light emitting element, 422, 4
24... Light receiving element, 421, 423, 431, 4
33... Grid, 425, 426, 435... Slit, 427... Fixed head adjustment direction. Figure 7:
61, 62, 71, 72, 81, 82...Reading circuit, 63...Rough measurement waveform forming circuit, bistable trigger circuit, 64...AND gate, 65...Angle measurement starting circuit, 66...AND gate, 67... Direction measurement starting circuit, 68... Addition/subtraction counter, 69... OR gate, 70... Rotation speed detection circuit, 710, 72
0,810,820...AND gate, 73,8
3...OR gate, 75...Micro measurement waveform forming circuit, 76...AND gate, 77...AND gate, 78, 79...divider, 80...AND gate, 74...clock circuit, 90... Coarse measurement counter, 91... Fine measurement counter, 92... Correction circuit,
93...Display start circuit, 94...AND gate,
95...Decoding/numerical display circuit.

Claims (1)

[Scope of Claims] 1 (i) The first indicators TF, TF0, TF180 and the second indicator 6 are arranged around a second rotating shaft arranged to substantially coincide with the axis of the first rotating shaft 7. , TM, TM 180 and a second indicator group including at least the third indicators 5, 50, each indicator rotates relative to the first indicator group. First indicators TF, TF0, related to the rotation axis 7
TF180 and second indicator 6, TM,
It belongs to either the first group containing TM180 or the second group containing at least one third indicator related to the second rotation axis, and further, the indicators of either group include: (ii) when the reference marks of the indicators of one group pass in front of the indicators of the other group; detecting coincident signals of two corresponding series; (iii) measuring the phase difference between the coincident signals of one series and the adjacent coincident signals of the other series; ,TF
0, TF180 and second indicator 6, TM,
A method of coding an angle defined by an angular displacement formed around the first rotation axis 7 between the TM 180 and the TM 180 into digital values, in which when measuring the phase difference, one series is matched. The total phase displacement obtained between the signal and the adjacent coincident signal of the other series was accumulated over one or several complete revolutions between the indicators of the first group and the indicators of the second group. A method for digitally coding an angle, comprising: averaging the reference marks to compensate for non-uniformity in the angular spacing of the reference marks so that the angle measurement is highly accurate. 2 A first indicator capable of relative movement to perform relative angular displacement about the first rotation axis 7
TF, TF0, TF180 and second indicator 6,
TM0, TM180; a first indicator group including a first indicator and a second indicator, centered on a second rotation axis arranged to substantially coincide with the axis of the first rotation axis; and at least one a device 4, 55 for causing a relative rotational movement between the indicator group and the second indicator group comprising a third indicator group 5, 50; belonging to a first group including one indicator TF, TF0, TF180 and a second indicator 6, TM0, TM180, or a second group including at least a third indicator 5, 50 related to the second rotation axis; One group of indicators includes
It includes a series of reference marks 51, 52 formed at equal intervals around a first axis of rotation; a device for detecting coincidence signals of two series; and an angle detection device 9 for detecting a delay between a coincidence signal of one series and an adjacent coincidence signal of the other series.
and; the angle detection device 9 comprises a device 23, 2315 for determining the time corresponding to one or more complete relative rotations between the indicators of the first group and the indicators of the second group; the total phase shift displacement obtained between the coincident signal and the adjacent coincident signal of the other series over one or several complete revolutions for a time defined by said device 23, 2315; A digital angle coder, characterized in that it comprises an accumulating and averaging device 23, 2426; the angle measurement is made highly accurate by compensating for non-uniformities in the angular spacing of the reference marks.