(19)
(11)EP 2 037 580 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
02.09.2015 Bulletin 2015/36

(21)Application number: 07745268.8

(22)Date of filing:  14.06.2007
(51)Int. Cl.: 
G01D 5/244  (2006.01)
H03M 1/08  (2006.01)
H03M 1/12  (2006.01)
H03K 5/1536  (2006.01)
(86)International application number:
PCT/JP2007/062011
(87)International publication number:
WO 2008/004422 (10.01.2008 Gazette  2008/02)

(54)

DIGITAL FILTER DEVICE, PHASE DETECTION DEVICE, POSITION DETECTION DEVICE, AD CONVERSION DEVICE, ZERO CROSS DETECTION DEVICE, AND DIGITAL FILTER PROGRAM

DIGITALFILTERANORDNUNG, PHASENDETEKTIONSANORDNUNG, POSITIONSDETEKTIONSANORDNUNG, AD-UMSETZUNGSANORDNUNG, NULLDURCHGANGS-DETEKTIONSANORDNUNG UND DIGITALFILTERPROGRAMM

DISPOSITIF DE FILTRE NUMÉRIQUE, DISPOSITIF DE DÉTECTION DE PHASE, DISPOSITIF DE DÉTECTION DE POSITION, DISPOSITIF DE CONVERSION A/N, DISPOSITIF DE DÉTECTION DU PASSAGE PAR ZÉRO ET PROGRAMME DE FILTRE NUMÉRIQUE


(84)Designated Contracting States:
DE FR IT

(30)Priority: 03.07.2006 JP 2006183817

(43)Date of publication of application:
18.03.2009 Bulletin 2009/12

(73)Proprietor: Toshiba Kikai Kabushiki Kaisha
Chiyoda-ku Tokyo 1008503 (JP)

(72)Inventors:
  • YOKOKAWA, Narutoshi
    Mishima-shi, Shizuoka 4110802 (JP)
  • SATO, Shouichi
    Izunokuni-shi, Shizuoka 4102321 (JP)

(74)Representative: 2K Patentanwälte Blasberg Kewitz & Reichel 
Partnerschaft mbB Schumannstrasse 27
60325 Frankfurt am Main
60325 Frankfurt am Main (DE)


(56)References cited: : 
DE-A1- 19 712 790
JP-A- 04 274 613
JP-A- S56 143 018
US-A- 4 926 072
US-A1- 2003 146 778
JP-A- 01 077 312
JP-A- 51 028 465
JP-A- 2005 002 926
US-A1- 2001 048 341
  
      
    Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


    Description

    TECHNICAL FIELD



    [0001] The present invention relates to a digital filter device, a phase detection device, a position detection device, an AD conversion device, a zero cross detection device, and a digital filter program.

    BACKGROUND ART



    [0002] Binary digital signals obtained by converting analog signals output from a resolver to a digital format (RD conversion, comparate) and other digital signals representing a waveform by fluctuation of the signal level (potential etc.) contain noise due to various factors mixed in them.

    [0003] FIG. 7A to FIG. 7E are diagrams explaining the effects of chattering upon a zero cross of an analog signal and a digital signal. FIG. 7A is a diagram showing 1 cycle's worth of an analog signal Sga1. The abscissa indicates the time (phase), while the ordinate indicates the signal level. Further, FIG. 7B is a diagram showing a digital signal Sgd1 obtained by converting the analog signal Sga1 of FIG. 7A to a digital format. The abscissa indicates the time (phase), while the ordinate indicates the signal level.

    [0004] As shown in FIG. 7A, the analog signal Sga1 is identified in its period and phase based on a zero cross point P1 crossing a signal level forming the standard (crossing zero), that is, the reference level Vo. Further, as shown in FIG. 7B, the position of the zero cross point of the analog signal Sga1 corresponds to the position of a rising edge Ed1 (or trailing edge) of the digital signal Sgd1.

    [0005] FIG. 7C is an enlarged diagram of a region R1 of FIG. 7A, that is, an enlarged diagram near the zero cross point P1, while FIG. 7D is an enlarged diagram of a region R2 of FIG. 7B, that is, an enlarged diagram near the rising edge Ed1. Note that, in FIG. 7D, a clock signal Sgc1 is shown as well.

    [0006] When viewing this macroscopically as in FIG. 7A and FIG. 7B, the analog signal Sga1 crosses zero at one point, while the digital signal Sgd1 has one rising edge Ed1. However, when viewing this microscopically as in FIG. 7C, due to chattering, the analog signal Sga1 crosses zero at a plurality of points (zero cross points P'1 to P'3). Further, as shown in FIG. 7D, if the period of the chattering is the period of the clock signal Sgc or more, the digital signal Sgd1 has a plurality of rising edges Ed'1 and Ed'2.

    [0007] FIG. 7E is a diagram explaining the effect of the chattering upon phase measurement and shows a binary digital signal Sgd3 obtained by converting a plurality of cycles' worth of the analog signal Sga1 to a digital format and a reference signal Sgd2 compared with the digital signal Sgd3. The abscissa indicates the time (phase), while the ordinate indicates the signal level.

    [0008] As shown at the left side of FIG. 7E on the page, when one rising edge should be generated in the digital signal Sgd3 corresponding to a rising edge Ed11 of the reference signal Sgd2, yet a plurality of rising edges Ed3 and Ed4 are generated due to chattering, the phase difference of the digital signal Sgd3 of the reference signal Sgd2 from the edge Ed11 will fluctuate (suffer from error) by exactly the phase difference between the rising edges Ed3 and Ed4.

    [0009] Further, as shown at the right side of FIG. 7E on the page, if a rising edge Ed6 is generated in the digital signal Sgd3 corresponding to a rising edge Ed12 of the reference signal Sgd2, a rising edge Ed5 will sometimes end up being generated due to chattering near a trailing edge of a half of a cycle before that rising edge Ed6. In this case, when the phase difference between the rising edge Ed12 and the rising edge Ed6 should be detected, the phase difference between the rising edge Ed12 and the rising edge Ed5 is liable to be erroneously detected and an error of half of a cycle is liable to occur.

    [0010] In order to solve such a problem, the technique using a hysteresis comparator is known (for example, Patent Document 1). In this technique, as shown in FIG. 7C, once the signal level of the analog signal Sga1 falls below the reference level V0 (crosses zero at the zero cross point P'1), when the signal level of the analog signal Sga1 exceeds a hysteresis level Vh, the next zero down cross will be detected by assuming that the analog signal Sga1 has crossed zero upward so as to thereby to remove the effect of chattering.

    [0011] Note that, although not concerning the technique of removing the effect of noise exerted upon the zero cross, in a system judging a combustion state of an internal combustion engine by detection of current flowing between electrodes of a spark plug, the technique of judging the effect of smoke upon the current flowing between electrodes of the spark plug based on the time during which the current flowing between electrodes of the spark plug exceeds a predetermined level or the like is known (Patent Documents 2 and 3).

    [0012] However, even if setting the hysteresis level as in Patent Document 1, if the amplitude of the noise produced due to chattering etc. exceeds the hysteresis level, ultimately the noise will end up appearing in the output digital signal as well. In particular, as explained above, in the case of identifying the phase difference, even if noise of a relatively fine time occurs, a relatively large error of half a cycle is liable to end up occurring.
    Patent Document 1: Japanese Patent Publication (A) No. 2004-12168
    Patent Document 2: Japanese Patent Publication (A) No. 2004-239085
    Patent Document 3: Japanese Patent Publication (A) No. 11-50941

    [0013] US 4 926 072 A relates to a circuit for eliminating noises contained in a signal. The level of a noise removed signal which is currently delivered is compared against the level of an input signal to be detected. When chattering arises, a counter is reset and counting is started from the initial value. Therefore, the time that the count reaches the predetermined value (= the time of inverting the signal level of the output signal) changes according to the change of the end time of chattering. A further counter is reset and counting is started from the initial value as well. The time that the count reaches the predetermined value (the end time of T1 in Fig. 4) changes according to the change of the start time or end time of chattering. Switching of a signal level of the output signal with the same delay from the trigger signal whether or not there is chatter on the trigger signal is not disclosed.

    DISCLOSURE OF THE INVENTION


    TECHNICAL PROBLEM



    [0014] It is an object of the present invention to provide an enhanced digital filter device and an enhanced phase detection device, position detection device, AD conversion device and zero cross detection device using such a digital filter device as well as a digital filter program for making a computer function as such a digital filter device.

    [0015] These problems are solved by a digital filter device according to claim 1 and 2, respectively, by a phase detection device according to claim 4, by a position detection device according to claim 5, by an AD conversion device according to claim 6, by a zero cross detection device according to claim 7 using such a digital filter device as well as a digital filter program according to claim 8 for making a computer function as such a digital filter device.

    TECHNICAL SOLUTION



    [0016] A digital filter device of a first aspect of the present invention is a digital filter device for filtering a binary input digital signal and outputting a binary output digital signal, comprising: a first accumulating portion configured to cumulatively add a time during which signal levels of the input digital signal and the output digital signal do not match so as to calculate a first accumulation time, suspend cumulatively adding and hold a time cumulatively added hitherto when the signal levels of the input digital signal and the output digital signal match, restart cumulatively adding from the time held when the signal levels of the input digital signal and the output digital signal do not match, and reset the first accumulation time to a predetermined first initial value and switch the time for cumulative addition between the time of mismatching of the signal levels and the time of matching of the signal levels when the first accumulation time has reached a predetermined first reference value; and a switch portion configured to switch the signal level of the output digital signal when the first accumulation time has reached the first reference value.

    [0017] Preferably, the digital filter device further including a status judgment portion configured to become a prohibit state that output of a predetermined second enable signal is prohibited when a predetermined clear signal has been input, the prohibit state is released conditional on at least the first enable signal being once input, and output the second enable signal while the first enable signal is not being input in the state where the prohibit state is released; and a second counter configured to count the clock signals while the second enable signal is being input so as to calculate a second count, and reset the second count to a predetermined second initial value and output the clear signal when the second count has reached a predetermined second reference value, wherein the first counter is configured to reset the first count to the first initial value when the clear signal is being input.

    [0018] A digital filter device of a second aspect of the present invention is a digital filter device for filtering a binary input digital signal and outputting a binary output digital signal, comprising: a first accumulating portion configured to cumulatively add a time during which signal levels of the input digital signal and the output digital signal match so as to calculate a first accumulation time, suspend cumulatively adding and hold a time cumulatively added hitherto when the signal levels of the input digital signal and the output digital signal do not match, restart cumulatively adding from the time held when the signal levels of the input digital signal and the output digital signal match, and reset the first accumulation time to a predetermined first initial value and switch the time for cumulative addition between the time of matching of the signal levels and the time of mismatching of the signal levels when the first accumulation time has reached a predetermined first reference value; and a switch portion configured to switch the signal level of the output digital signal when the first accumulation time has reached the first reference value.

    [0019] Preferably, the digital filter device further including a second accumulation portion configured to cumulatively add the time of the second level in a case where the time cumulatively added in the first accumulation portion is the time of the first level and cumulatively add the time of the first level in a case where the time cumulatively added in the first accumulation portion is the time of the second level so as to calculate a second accumulation time, reset the second accumulation time to a predetermined second initial value and prohibit the calculation of the second accumulation time when the second accumulation time has reached a predetermined second reference value, and release the prohibition on the calculation of the second accumulation time conditional on at least the first accumulation time being once calculated at the first accumulation portion, wherein the first accumulation portion is configured to reset the first accumulation time to the first initial value when the second accumulation time has reached a predetermined second reference value.

    [0020] A phase detection device of a third aspect of the present invention is a phase detection device comprising a digital filter device as outlined above; and a phase detection portion detecting a phase of the input digital signal based on the output digital signal output from the digital filter device.

    [0021] A position detection device of a fourth aspect of the present invention is a position detection device a resolver shifting a phase of an input analog type waveform shaped excitation signal by exactly an amount in accordance with a rotation position of a resolver shaft and outputting the result; a comparator converting the signal output from the resolver to a binary input digital signal; a digital filter device as outlined above; and a position detection portion calculating a rotation position of the resolver based on a phase difference between a binary digital type reference signal having the same period and phase as those of the excitation signal and the output digital signal output from the digital filter device.

    [0022] An AD conversion device of a fifth aspect of the present invention is an AD conversion device for converting an analog signal to a binary digital signal, comprising: the comparator to AD convert the analog signal to the binary digital signal for output with reference to a reference level; and a digital filter device as outlined above.

    [0023] A zero cross detection device of a sixth aspect of the present invention is a zero cross detection device in the digital filter device as outlined above, wherein the zero cross detection device is further configured to output a zero cross detection signal when the accumulated time has reached a predetermined reference value.

    [0024] A digital filter program of a seventh aspect of the present invention is a digital filter program for making a computer function function as a digital filter device as outlined above.

    EFFECT OF THE INVENTION



    [0025] According to the present invention, the effect of noise can be removed from the edge of a digital signal.

    BRIEF DESCRIPTION OF THE DRAWINGS



    [0026] 

    [FIG. 1] A block diagram schematically showing the overall configuration of a rotation position detection device according to an embodiment of the present invention.

    [FIG. 2] A block diagram showing the configuration of a digital filter device of the rotation position detection device of FIG. 1.

    [FIG. 3] A timing chart explaining an operation of the digital filter device of FIG. 2 in a case where there is no noise.

    [FIG. 4] A timing chart explaining the operation of the digital filter device of FIG. 2 in a case where there is a relatively small noise.

    [FIG. 5] A diagram explaining an effect of the digital filter device of FIG. 2.

    [FIG. 6] A timing chart explaining the operation of the digital filter device of FIG. 2 in a case where there is a relatively large noise.

    [FIG. 7] A diagram explaining the problem of the prior art.

    [FIG. 8] A block diagram explaining a modification of the present invention.


    EXPLANATION OF NOTATIONS



    [0027] 4... digital filter device, 12... toggle flip-flop (first accumulation portion, switch portion), 13... XOR circuit (first accumulation portion), 14... charge counter (first accumulation portion), Sdi, DIN... input digital signals, and Sdo, DOUT... output digital signals.

    BEST MODE FOR CARRYING OUT THE INVENTION



    [0028] FIG. 1 is a block diagram schematically showing the overall configuration of a rotation position detection device 1 according to an embodiment of the present invention. The rotation position detection device 1 is provided with a resolver 2 attached to a motor or other detected object, a comparator 3 which performs predetermined processing with respect to an analog signal Sa from the resolver 2, a digital filter device 4, and a phase difference detection portion 5 and is configured as the device which identifies a rotation position of the detected object. The operation thereof is as follows.

    [0029] The resolver 2 receives as input the two phases of analog type excitation signals Ss and Sc produced by an excitation circuit 6 based on a clock signal CLK from a clock generator. The signal levels of the excitation signals Ss and Sc respectively fluctuate by sin(ωt) and cos(ωt). The resolver 2 outputs the analog signal Sa obtained by shifting the input excitation signal Ss by exactly a rotation angle θ of the resolver shaft. The signal level of the analog signal Sa fluctuates by sin(ωt+θ).

    [0030] The comparator 3, as explained with reference to FIG. 7, while synchronized with the clock signal CLK from the clock generator, outputs a signal which becomes constant at a predetermined low level (or high level) of the signal level when the signal level of the analog signal Sa is high relative to the predetermined reference level and outputs a signal which becomes constant at a predetermined high level (or low level) of the signal level when the signal level of the analog signal Sa is low relative to the predetermined reference level. Namely, it AD converts the analog signal Sa to a binary digital signal Sdi for output with reference to the reference level.

    [0031] The digital filter device 4 outputs a binary digital signal Sdo obtained by filtering the digital signal Sdi from the comparator 3. Namely, the digital filter device 4 removes noise due to chattering from the digital signal Sdi and outputs the result.

    [0032] The phase difference detection portion 5 receives as input the digital signal Sdo output from the digital filter device 4 and a reference signal Sds output from a reference signal generation portion 8. The reference signal Sds is a digital signal having the same period and phase as those of the excitation signal Ss output from the excitation circuit 6. Further, the phase difference detection portion 5 detects the phase difference between the reference signal Sds and the digital signal Sdo. This detection of the phase difference is equivalent to the detection of the phase difference between the excitation signal Ss and the analog signal Sa. The rotation angle θ (rotation position) of the resolver shaft is identified by this.

    [0033] FIG. 2 is a block diagram showing the configuration of the digital filter device 4. The digital filter device 4 is configured as a device which filters an input signal DIN and outputs an output signal DOUT. Note that, as shown in FIG. 1, in a case where the digital filter device 4 is assembled in the rotation position detection device 1, the input signal DIN equals the digital signal Sdi, and the output signal DOUT equals the digital signal Sdo. All flip-flops and counters explained later receive as input clock signals CLK from the clock generator, that is, the same clock signals. However, the illustration is omitted in FIG. 2. In the following explanation, an explanation will be sometimes given assuming the signal level of the digital signal to be the value of 1 bit of information, that is, 0 (false, low level) or 1 (true, high level). Further, an explanation will be sometimes given assuming a change of the value from 0 to 1 as the rising edge of the signal level and a change from 1 to 0 as the trailing edge of the signal level.

    [0034] A delay flip-flop (DFF) 11 receives as input the input signal DIN and the clock signal CLK, holds the value of the input signal DIN at a point of time of the rising edge (or trailing edge) of the clock signal CLK, and outputs that value from an output terminal Q in synchronization with the next clock signal CLK. A toggle flip-flop (toggle FF) 12 outputs the output signal DOUT and alternately switches the value of the output signal DOUT between 1 and 0 each time a carry on signal ON_RCO output from a charge counter 14 as the trigger signal is input. An exclusive OR logic circuit (XOR circuit: eXclusive OR circuit) 13 receives as input the output signal from the DFF 11 and the output signal from the toggle FF 12 and outputs a signal having a value of 1 as a first enable signal EN1 when the values of the two signals do not match.

    [0035] The charge counter 14 performs counting in synchronization with the clock signal CLK while the first enable signal EN1 is being input. The counting is carried out within a range of n bits of information. Namely, the counting is carried out 2n times from the initial value (for example 0) until the count becomes the upper limit value (for example 2n) and is cleared (reset to the initial value) when the count has reached the upper limit value. Further, the count is cleared at a time when a carry off signal OFF_RCO output from a stay counter 19 as a clear signal is input to the charge counter 14. Note that, since the counting is carried out in synchronization with the clock signal CLK, the charge counter 14 cumulatively adds the time (cycles of clock signal) while the first enable signal EN1 is being input. When the count has reached the upper limit value, the charge counter 14 outputs the carry signal ON_RCO having the value of 1 for exactly a time of 1 clock.

    [0036] An edge detection flip-flop (edge detection FF) 15 receives as input the signal from the XOR circuit 13 and outputs a signal having the value of 1 when detecting a trailing edge of the signal from the XOR circuit 13. A latch flip-flop (latch FF) 16 receives as input the signal from the edge detection FF15 and the clock signal CLK, holds the signal from the edge detection FF15 when the clock signal CLK is 1, and outputs the same. An inverter circuit (NOT: NOT circuit) 17 inverts the signal from the XOR circuit 13 for output. An AND circuit 18 outputs an AND logic of the signal from the latch FF 16 and the signal from the inverter circuit 17.

    [0037] The stay counter 19 performs the counting in synchronization with the clock signal CLK while a signal having the value of 1 is being input from the AND circuit 18 as a second enable signal EN2. The counting is carried out within a range of m bits of information. Namely, the counting is carried out 2m times from the initial value (for example 0) until the count becomes the upper limit value (for example 2m) and is cleared (reset to the initial value) when the count has reached the upper limit value. Further, the count is cleared at a time when a carry on signal ON_RCO output from the charge counter 14 as the clear signal is input to the stay counter 19. Note that, since the counting is carried out in synchronization with the clock signal CLK, the stay counter 19 cumulatively adds the time (cycles of the clock signal) while the second enable signal EN2 is being input. The stay counter 19 outputs the carry OFF signal OFF_RCO having the value of 1 for exactly the time of one clock when the count has reached the upper limit value.

    [0038] The operation of the digital filter device 4 having the above configuration will be explained next.

    [0039] First, a case where there is no chattering noise or other noise will be explained. FIG. 3 is a timing chart showing values of output signals of the different portions when there is no noise.

    [0040] The initial value of the signal output by the toggle FF 12 and the initial value of the signal DIN input to the DFF 11 (the signal output from the DFF 11) are given the same value (time t0). For example, these are made 0. Namely, an input signal DIN having the value of 0 is input to the digital filter device 4, and the digital filter device 4 outputs an output signal DOUT of the value 0.

    [0041] When the input signal DIN changes from 0 to 1 (time t1), the output of the XOR circuit 13 changes from 0 to 1. Namely, the first enable signal EN1 is output from the XOR circuit 13. While the first enable signal EN1 is being input, the charge counter 14 repeats the counting in synchronization with the clock signal CLK and outputs the carry on signal ON_RCO for exactly the time of one clock at a time when the count has reached the upper limit value (time t2). When the carry on signal ON_RCO is input as the trigger signal to the toggle FF 12, the toggle FF 12 changes the output from 0 to 1. As a result, the value of the signal DOUT becomes 1. Further, when the carry on signal ON_RCO is input as a clear signal to a clear terminal CLR of the stay counter 19, the stay counter 19 resets the count to the initial value.

    [0042] When the output of the toggle FF 12 changes from 0 to 1, the output of the XOR circuit 13 changes from 1 to 0. Due to this, the outputs of the edge detection FF 15 and latch FF 16 and the inverter circuit 17 become 1, and the output of the AND circuit 18 becomes 1. Namely, the second enable signal EN2 is output. The stay counter 19 repeats the counting in synchronization with the clock signal CLK while the second enable signal EN2 is being input and outputs the carry off signal OFF_RCO for exactly the time of one clock at a time when the count has reached the upper limit value (time t3). By this signal, the edge detection FF 15, latch FF 16, and charge counter 14 are reset to the initial state.

    [0043] After that, until the input signal DIN changes from 1 to 0, the flip-flops 11, 12, 15, and 16 and counters 14 and 19 do not operate. When the input signal DIN changes from 1 to 0 (time t4), these portions operate in the same way as the time when it changed from 0 to 1.

    [0044] As described above, when there is no chattering noise or other noise, the digital filter device 4 outputs the output signal DOUT having the same value as that of the input signal DIN with a delay from the input of the input signal DIN of exactly a time T1 during which the charge counter 14 performs a full count (time from t1 to t2, cycle of clock x 2n). Note that, the phase difference detection portion 5 performs correction by subtracting a phase corresponding to the time T1 from the phase difference between the output signal DOUT and the reference signal Sds and identifies the rotation position.

    [0045] Next, a case where chattering or other noise is generated will be explained. FIG. 4 is a timing chart showing values of output signals of the different portions in a case where relatively minute noise represented by chattering is generated.

    [0046] The times t10 to t11 are the same as the times t0 to t1 of FIG. 3. The charge counter 14 starts counting from the time t11. When the output signal of the input signal DIN changes from 1 to 0 before the count of the charge counter 14 has reached the upper limit value (time t12), the output of the XOR circuit 13 changes from 1 to 0. Accordingly, an edge is detected by the edge detection FF 15, the output of the latch FF is fixed to 1, the output of the inverter circuit 17 becomes 1, the second enable signal EN2 is output from the AND circuit 18, and the stay counter 19 starts counting. Further, the charge counter 14 suspends counting and holds the value counted hitherto.

    [0047] After that, when the input signal DIN changes from 0 to 1 again (time t13), the output of the AND circuit 18 becomes 0. Accordingly, the stay counter 19 suspends counting and holds the value counted hitherto. On the other hand, the charge counter 14 restarts counting. Further, when the input signal DIN changes from 1 to 0 again (time t14), the charge counter 14 suspends counting, and the stay counter 19 restarts counting. Namely, the charge counter 14 performs counting in the state where the input signal DIN is 1 (state where the value of the input signal DIN and the value of the output signal DOUT are different), while the stay counter 19 performs counting in a state where the input signal DIN is 0 (state where the value of the input signal DIN and the value of the output signal DOUT are the same).

    [0048] When the charge counter 14 reaches the upper limit value earlier than the stay counter 19, in the same way as the time t2 of FIG. 3, the carry on signal ON_RCO is output from the charge counter 14, and the toggle FF 12 changes the output signal DOUT from 0 to 1 (time t16) . Namely, the digital filter device 4 outputs an output signal DOUT having the same value as that of the input signal DIN with a delay from the input of the input signal DIN of exactly a time T2 required for the charge counter 14 to perform a full count (time from t11 to t16, larger than clock period x 2n, but smaller than clock period x 2n+m). Note that, a case where the stay counter 19 has reached the upper limit value earlier than the charge counter 14 will be explained later.

    [0049] The stay counter 19 receives as input the carry on signal ON_RCO output from the charge counter 14 and, after the count is reset to the initial value (after the time t16), performs counting in the same way as that after the time t2 of FIG. 3. Further, it outputs the carry off signal OFF_RCO for exactly the time of one clock at the time when the count has reached the upper limit value (time t17). By the resetting of the edge detection FF 15 and the latch FF 16 to the initial state by this signal, the stay counter 19 does not perform counting even in the state where the value of the input signal DIN and the value of the output signal DOUT are the same. The count of the stay counter 19 is maintained at the initial value until the counting is started next at the charge counter 14.

    [0050] FIG. 5 is a diagram explaining the advantageous effect of the digital filter device 4, in which FIG. 5A shows a timing chart when chattering does not occur, and FIG. 5B shows a timing chart when chattering occurs.

    [0051] As shown in FIG. 5A, when there is no chattering, the digital signal DIN obtained by comparing the analog signal Sa has a rising edge Ed21 at schematically the same point of time as the zero cross point P11 in the case when viewing the analog signal Sa macroscopically (in the diagram, indicated by an approximation line). Further, the digital signal DOUT output from the digital filter device 4 has a rising edge Ed22 corresponding to the rising edge Ed21 with a delay of the time T1 (clock period x 2n) from the rising edge Ed21 of the digital signal DIN.

    [0052] On the other hand, as shown in FIG. 5B, when there is chattering, the digital signal DIN obtained by comparing the analog signal Sa has a plurality of rising edges at the same points of time as the plurality of zero cross points in the case when viewing the analog signal Sa microscopically. However, the digital signal DOUT output from the digital filter device 4 has only a rising edge Ed31 corresponding to the macroscopic zero cross point P11 with a delay of exactly the time T2 (larger than clock period x 2n, smaller than clock period x 2n+m) from the first rising edge Ed25 among a plurality of rising edges of the digital signal DIN.

    [0053] Accordingly, even when a plurality of edges are produced due to chattering in the digital signal DIN after comparison, only one edge corresponding to the zero cross point P11 seen macroscopically is generated.

    [0054] Further, a plurality of edges of the digital signal DOUT are generated at both points of time before the zero cross point P11 and points of time after it. Therefore, if n (parameter for counting time) is adequately set in accordance with the characteristic etc. of chattering of each device so that the time during which the charge counter 14 performs counting at the points of time before the zero cross point P11 (T4+T5+T6) and the time during which the charge counter 14 does not perform counting at the points of time after the zero cross point P11 (T7+T8) become the same degree, it is possible to make a time T3 from the zero cross point P11 up to the rising edge Ed31 and a time T1 from the zero cross point P11 up to the rising edge Ed22 in the case where there is no chattering the same degree.

    [0055] FIG. 6 is a timing chart showing values of output signals of portions when the stay counter 19 has reached the upper limit value earlier in the case where the charge counter 14 and the stay counter 19 alternately perform counting.

    [0056] The time t20 to time t24 are the same as time t10 to time t15 of FIG. 4. At the time t25, when the stay counter 19 has reached the upper limit value earlier than the charge counter 14, the carry off signal OFF_RCO is output from the stay counter 19, and the charge counter 14, edge detection FF 15, and latch FF 16 are reset to the initial state. Further, the toggle FF 12 does not change the output. Namely, the digital filter device 4 judges that the change of the input signal DIN is not due to a zero cross or noise related to a zero cross such as chattering noise and enters a standby state for generating a new edge. Due to this, the effect of any relatively large noise generated irrelevant to a zero cross is removed.

    [0057] Note that, in the embodiment described above, the charge counter 14 is an example of the first counter of the present invention, the XOR circuit 13 is an example of the level judgment portion of the present invention, the stay counter 19 is an example of the second counter of the present invention, the combination of the edge detection FF 15, latch FF 16, inverter circuit 17, and AND circuit 18 is an example of the status judgment portion of the present invention, the combination of the toggle FF 12, XOR circuit 13, and charge counter 14 is an example of the first accumulation portion of the present invention, the toggle FF 12 is an example of the switch portion of the present invention, the combination of the edge detection FF 15, latch FF 16, inverter circuit 17, AND circuit 18, and stay counter 19 is an example of the second accumulation portion of the present invention, the phase difference detection portion 5 is an example of the phase detection portion and position detection portion of the present invention, the rotation position detection device 1 is an example of the phase detection device of the present invention, the combination of the comparator 3 and digital filter device 4 is an example of the AD conversion device of the present invention, and the charge counter 14 is an example of the zero cross detection device of the present invention.

    [0058] The portions shown in FIG. 1 and FIG. 2 may be configured by logic circuits or constructed in a computer including a CPU, RAM, ROM, external memory device etc. by the CPU running a program stored in the ROM etc. When the digital filter device 4 is constructed by the CPU running a program, the program is an example of the digital filter program of the present invention.

    [0059] The present invention is not limited to the above embodiments and may be carried out in various ways.

    [0060] The input signal filtered by the digital filter device is not limited to one obtained by conversion of an analog signal to a digital signal and may be one generated (detected) as a digital signal from the start as well.

    [0061] The input signal and the output signal of the digital filter device may differ in signal level as well. FIG. 8 shows a modification of the digital filter device which inverts the value of the input binary digital signal DIN for output. In this modification, an inverter circuit 51 is provided immediately after the toggle FF 12, and a coincidence circuit 52 (an example of the level judgment portion) which outputs the first enable signal EN1 when values of the two input digital signals match is provided in place of the XOR circuit 13. The coincidence circuit 52 is provided with, for example, an AND circuit 53 receiving as input the output signals of the inverter circuit 51 and DFF 11, inverter circuits 54 and 55 which invert output signals of the inverter circuit 51 and DFF 11, an AND circuit 56 receiving as input the output signals of the inverter circuits 54 and 55, and an OR circuit 57 receiving as input the output signals of the AND circuit 53 and AND circuit 56.

    [0062] The phase detection portion and the phase detection device are not limited to the phase difference detection portion 5 and the rotation position detection device 1 which detect a phase difference between the reference signal of the resolver and the output signal and detect the rotation position on a basis of that phase difference. For example, these may be a phase detection portion and a phase detection device designed for the detection of the phase itself of a measured object which rotates or rocks.

    [0063] The zero cross detection device may be one detecting the zero cross point (edge) of a signal representing a waveform. Accordingly, the input signal for which the zero cross is detected may be an analog signal or a digital signal. Further, the detection result of the zero cross point can be utilized for various objects. For example, it can be utilized for a device which does not have to output the digital signal after removing the noise (the output signal DOUT of the embodiment), for example, a wave measurement device which calculates a number of waves and frequencies of waves on a basis of the zero cross points of the analog signal obtained by measuring a wave height. Even in the case where the digital signal from which noise was removed is output, the output signal is not limited to a binary digital signal. Predetermined correction processing based on the detection of the zero cross point may be carried out with respect to the input signal to output an analog signal or output a digital signal having a signal level fluctuating at a plurality of levels.


    Claims

    1. A digital filter device (4) for filtering a binary input digital signal and outputting a binary output digital signal, comprising:

    a first accumulating portion (14) configured to
    cumulatively add a time during which signal levels of the input digital signal and the output digital signal do not match so as to calculate a first accumulation time,
    suspend cumulatively adding and hold a time cumulatively added hitherto when the signal levels of the input digital signal and the output digital signal match,
    restart cumulatively adding from the time held when the signal levels of the input digital signal and the output digital signal do not match, and
    reset the first accumulation time to a predetermined first initial value and switch the time for cumulative addition between the time of mismatching of the signal levels and the time of matching of the signal levels when the first accumulation time has reached a predetermined first reference value; and

    a switch portion (12) configured to switch the signal level of the output digital signal when the first accumulation time has reached the first reference value.


     
    2. A digital filter device (4) for filtering a binary input digital signal and outputting a binary output digital signal, comprising:

    a first accumulating portion (14) configured to
    cumulatively add a time during which signal levels of the input digital signal and the output digital signal match so as to calculate a first accumulation time,
    suspend cumulatively adding and hold a time cumulatively added hitherto when the signal levels of the input digital signal and the output digital signal do not match,
    restart cumulatively adding from the time held when the signal levels of the input digital signal and the output digital signal match, and
    reset the first accumulation time to a predetermined first initial value and switch the time for cumulative addition between the time of matching of the signal levels and the time of mismatching of the signal levels when the first accumulation time has reached a predetermined first reference value; and

    a switch portion (12) configured to switch the signal level of the output digital signal when the first accumulation time has reached the first reference value.


     
    3. A digital filter device as set forth in clam 1 or 2, further comprising:

    a second accumulation portion (19) configured to
    cumulatively add the time of the second level in a case where the time cumulatively added in the first accumulation portion is the time of the first level and cumulatively add the time of the first level in a case where the time cumulatively added in the first accumulation portion is the time of the second level so as to calculate a second accumulation time,
    reset the second accumulation time to a predetermined second initial value and prohibit the calculation of the second accumulation time when the second accumulation time has reached a predetermined second reference value, and
    release the prohibition on the calculation of the second accumulation time conditional on at least the first accumulation time being once calculated at the first accumulation portion, wherein

    the first accumulation portion is configured to reset the first accumulation time to the first initial value when the second accumulation time has reached a predetermined second reference value.


     
    4. A phase detection device (1) comprising:

    a digital filter device (4) as set forth in clam 1 or 2; and

    a phase detection portion (5) detecting a phase of the input digital signal based on the output digital signal output from the digital filter device.


     
    5. A position detection device comprising:

    a resolver (2) shifting a phase of an input analog type waveform shaped excitation signal by exactly an amount in accordance with a rotation position of a resolver shaft and outputting the result;

    a comparator (3) converting the signal output from the resolver to a binary input digital signal;

    a digital filter device (4) as set forth in clam 1 or 2; and

    a position detection portion (5) calculating a rotation position of the resolver based on a phase difference between a binary digital type reference signal having the same period and phase as those of the excitation signal and the output digital signal output from the digital filter device.


     
    6. An AD conversion device for converting an analog signal to a binary digital signal, comprising:

    the comparator (3) to AD convert the analog signal (Sa) to the binary digital signal (Sdi) for output with reference to a reference level;

    a digital filter device (4) as set forth in clam 1 or 2.


     
    7. A zero cross detection device (14) in the digital filter device (4) as set forth in clam 1 or 2, said zero cross detection device further configured to
    output a zero cross detection signal when the accumulated time has reached a predetermined reference value.
     
    8. A digital filter program for making a computer function as a digital filter device as set forth in clam 1 or 2.
     


    Ansprüche

    1. Digitale Filtereinrichtung (4) zum Filtern eines binären digitalen Eingangssignals und zum Ausgeben eines binären digitalen Ausgangsignals, umfassend:

    einen ersten Akkumulatorabschnitt (14) bzw. Abschnitt zum Sammeln von Ergebnissen fortlaufender Additionen, der ausgelegt ist,

    um eine Gesamtzeit fortlaufend aufzuaddieren, während der Signalpegel des digitalen Eingangssignals und des digitalen Ausgangsignals nicht miteinander übereinstimmen, um so eine erste Gesamtzeit zu berechnen,

    um das fortlaufende Aufaddieren der Gesamtzeit auszusetzen und eine bis dahin aufaddierte Gesamtzeit zu halten, wenn die Signalpegel des digitalen Eingangssignals und des digitalen Ausgangsignals miteinander übereinstimmen,

    um ein fortlaufendes Aufaddieren von der gehaltenen Zeit ab erneut zu beginnen, wenn die Signalpegel des digitalen Eingangssignals und des digitalen Ausgangsignals nicht miteinander übereinstimmen, und

    um die erste aufaddierte Gesamtzeit auf einen vorbestimmten ersten Anfangswert zurückzusetzen und den Zeitpunkt zum fortlaufenden Aufaddieren zwischen dem Zeitpunkt der Nichtübereinstimmung der Signalpegel und dem Zeitpunkt der Übereinstimmung der Signalpegel zu ändern, wenn die erste aufaddierte Gesamtzeit einen vorbestimmten ersten Referenzwert erreicht hat; und

    einen Schaltabschnitt (12), der ausgelegt ist, um den Signalpegel des digitalen Ausgangssignals umzuschalten, wenn die erste aufaddierte Gesamtzeit den ersten Referenzwert erreicht hat.


     
    2. Digitale Filtereinrichtung (4) zum Filtern eines binären digitalen Eingangssignals und zum Ausgeben eines binären digitalen Ausgangsignals, umfassend:

    einen ersten Akkumulatorabschnitt (14) bzw. Abschnitt zum Sammeln von Ergebnissen fortlaufender Additionen, der ausgelegt ist,

    um eine Gesamtzeit fortlaufend aufzuaddieren, während der Signalpegel des digitalen Eingangssignals und des digitalen Ausgangsignals nicht miteinander übereinstimmen, um so eine erste Gesamtzeit zu berechnen,

    um das fortlaufende Aufaddieren der Gesamtzeit auszusetzen und eine bis dahin aufaddierte Gesamtzeit zu halten, wenn die Signalpegel des digitalen Eingangssignals und des digitalen Ausgangsignals nicht miteinander übereinstimmen,

    um das fortlaufende Aufaddieren von dem gehaltenen Zeitpunkt ab erneut zu beginnen, wenn die Signalpegel des digitalen Eingangssignals und des digitalen Ausgangsignals miteinander übereinstimmen, und

    um die erste aufaddierte Gesamtzeit auf einen vorbestimmten ersten Anfangswert zurückzusetzen und den Zeitpunkt zum fortlaufenden Aufaddieren zwischen dem Zeitpunkt der Übereinstimmung der Signalpegel und dem Zeitpunkt der Nichtübereinstimmung der Signalpegel zu ändern, wenn die erste aufaddierte Gesamtzeit einen vorbestimmten ersten Referenzwert erreicht hat; und

    einen Schaltabschnitt (12), der ausgelegt ist, um den Signalpegel des digitalen Ausgangssignals umzuschalten, wenn die erste aufaddierte Gesamtzeit den ersten Referenzwert erreicht hat.


     
    3. Digitale Filtereinrichtung nach Anspruch 1 oder 2, weiterhin umfassend:

    einen zweiten Akkumulatorabschnitt (19) bzw. Abschnitt zum Sammeln von Ergebnissen fortlaufender Additionen, der ausgelegt ist

    um die Zeit des zweiten Pegels für den Fall fortlaufend aufzuaddieren, wenn die in dem ersten Akkumulatorabschnitt fortlaufende aufaddierte Gesamtzeit die Zeit des ersten Pegels ist, und um die Zeit des ersten Pegels für den Fall fortlaufend aufzuaddieren, wenn die in dem ersten Akkumulatorabschnitt fortlaufend aufaddierte Gesamtzeit die Zeit des zweiten Pegels ist, um eine zweite Gesamtzeit zu berechnen,

    um die zweite Gesamtzeit auf einen vorbestimmten zweiten Anfangswert zurückzusetzen und die Berechnung der zweiten Gesamtzeit zu verhindern, wenn die zweite Gesamtzeit einen vorbestimmten zweiten Referenzwert erreicht hat, und

    um das Verhindern der Berechnung der zweiten Gesamtzeit in Abhängigkeit von zumindest der ersten Gesamtzeit, die einmal von dem ersten Akkumulatorabschnitt berechnet worden ist, freizugeben, wobei

    der erste Akkumulatorabschnitt ausgelegt ist, um die erste Gesamtzeit auf den ersten Anfangswert zurückzusetzen, wenn die zweite Gesamtzeit einen vorbestimmten zweiten Referenzwert erreicht hat.


     
    4. Phasenerfassungseinrichtung (1), umfassend:

    eine digitale Filtereinrichtung (4) nach Anspruch 1 oder 2; und

    einen Phasenerfassungsabschnitt (5), der eine Phase des digitalen Eingangssignals auf der Grundlage des digitalen Ausgangssignals erfasst, das von der digitalen Filtereinrichtung ausgegeben wird.


     
    5. Positionserfassungsvorrichtung, umfassend:

    einen Drehmelder (2), die eine Phase eines analogen Eingangs-Anregungssignals exakt um einen Betrag in Übereinstimmung mit einer Drehstellung einer Drehmelder-Welle verschiebt und das Ergebnis ausgibt;

    einen Komparator (3), der das Ausgangssignal von dem Drehmelder in ein binäres digitales Eingangssignal umwandelt;

    eine digitale Filtereinrichtung (4) nach Anspruch 1 oder 2; und

    einen Positionserfassungsabschnitt (5), der eine Drehstellung des Drehmelders auf der Grundlage einer Phasendifferenz zwischen einem binären digitalen Referenzsignal mit der gleichen Periode und Phase wie das Anregungssignal und dem digitalen Ausgangsignal berechnet, das von der digitalen Filtereinrichtung ausgegeben wird.


     
    6. AD-Wandlungsvorrichtung zum Umwandeln eines analogen Signals in ein binäres digitales Signal, umfassend:

    den Komparator (3) zum AD-Wandeln das analogen Signals (Sa) in das binäre digitale Signal (Sdi) zur Ausgabe mit Bezug auf einen Referenzpegel;

    eine digitale Filtereinrichtung (4) nach Anspruch 1 oder 2.


     
    7. Nulldurchgangs-Erfassungseinrichtung (14) in der digitalen Filtervorrichtung (4) nach Anspruch 1 oder 2, wobei die Nulldurchgangs-Erfassungseinrichtung weiterhin ausgelegt ist
    um ein Nulldurchgangs-Erfassungssignal auszugeben, wenn die fortlaufend aufaddierte Gesamtzeit einen vorbestimmten Referenzwert erreicht hat.
     
    8. Digitales Filterprogramm zur Bereitstellung einer Computerfunktion als digitale Filtereinrichtung nach Anspruch 1 oder 2.
     


    Revendications

    1. Un dispositif de filtrage numérique (4) pour filtrer un signal numérique binaire d'entrée et pour générer un signal numérique binaire de sortie, comprenant :

    une première partie accumulatrice (14) configurée pour
    accumuler un temps durant lequel des niveaux de signaux du signal numérique d'entrée et du signal numérique de sortie ne correspondent pas de manière à calculer une première durée d'accumulation,
    suspendre l'accumulation et maintenir une durée accumulée dès lors que les niveaux de signal du signal numérique d'entrée et du signal numérique de sortie correspondent,
    redémarrer l'accumulation à partir du temps maintenu dès lors que les niveaux de signal du signal numérique d'entrée et du signal numérique de sortie ne correspondent plus, et
    réinitialiser la première durée accumulée à une première valeur initiale prédéterminée et commuter le temps pour l'accumulation entre le moment de non correspondance des niveaux de signaux et le moment de correspondance des niveaux de signaux lorsque la première durée d'accumulation a atteint une première valeur de référence ; et

    une partie de commutation (12) configurée pour commuter le niveau de signal du signal numérique de sortie lorsque la première durée d'accumulation a atteint la première valeur de référence.


     
    2. Un dispositif de filtre numérique (4) pour filtrer un signal numérique binaire d'entrée et générer un signal numérique binaire de sortie, comprenant :

    une première partie accumulatrice (14) configurée pour
    accumuler un temps durant lequel des niveaux de signaux du signal numérique d'entrée et du signal numérique de sortie correspondent de manière à calculer une première durée d'accumulation,
    suspendre l'accumulation et maintenir une durée accumulée dès lors que les niveaux de signal du signal numérique d'entrée et du signal numérique de sortie ne correspondent pas,
    redémarrer l'accumulation à partir du temps maintenu dès lors que les niveaux de signal du signal numérique d'entrée et du signal numérique de sortie correspondent, et
    réinitialiser la première durée accumulée à une première valeur initiale prédéterminée et commuter le temps pour l'accumulation entre le moment de correspondance des niveaux de signaux et le moment de non-correspondance des niveaux de signaux lorsque la première durée d'accumulation a atteint une première valeur de référence ; et

    une partie de commutation (12) configurée pour commuter le niveau de signal du signal numérique de sortie lorsque la première durée d'accumulation a atteint la première valeur de référence.


     
    3. Un dispositif de filtrage numérique tel que défini dans la revendication 1 ou 2, comprenant en outre :

    une seconde partie accumulatrice (19) configurée pour
    accumuler le temps du second niveau dans une cas où la durée accumulée dans la première partie accumulatrice est la durée du premier niveau et accumuler le temps du premier niveau dans un cas où la durée accumulée dans la première partie accumulatrice est le temps du second niveau de manière à calculer une seconde durée d'accumulation,
    réinitialiser la seconde durée d'accumulation à une seconde valeur initiale prédéterminée et empêcher le calcul de la seconde durée accumulée lorsque la seconde durée accumulée atteint une seconde valeur de référence prédéterminée, et
    relâcher l'empêchement sur le calcul de la seconde durée accumulée sous la condition au moins que la première durée accumulée ait été calculée au niveau de la première partie accumulatrice ; dans laquelle

    la première partie accumulatrice est configurée pour réinitialiser la première durée d'accumulation à la première valeur initiale lorsque la seconde durée d'accumulation a atteint une seconde valeur de référence prédéterminée.


     
    4. Un dispositif de détection de phase (1) comprenant :

    un dispositif de filtrage numérique (4) tel que défini dans la revendication 1 ou 2; et

    une partie de détection de phase (5) détectant une phase du signal numérique d'entrée sur la base du signal numérique de sortie générée en sortie du dispositif de filtrage numérique.


     
    5. Un dispositif de détection de position comprenant :

    un résolveur (2) décalant une phase d'une forme d'onde analogique d'un signal d'excitation en forme d'onde d'entrée analogique d'une valeur correspondant à une position de rotation d'un arbre de résolveur et générant le résultat ;

    un comparateur (3) convertissant la sortie du signal du résolveur en un signal numérique binaire d'entrée ;

    un dispositif de filtrage numérique (4) tel que défini dans la revendication 1 ou 2; et

    une partie de détection de position (5) calculant une position de rotation du résolveur sur la base d'une différence de phase entre un signal de référence de type
    numérique binaire ayant les mêmes période et phase que celle du signal d'excitation et du signal numérique de sortie du dispositif de filtrage numérique.


     
    6. Un dispositif de conversion NA pour la conversion d'un signal analogique en un signal numérique, comprenant :

    le comparateur (3) pour la conversion AD du signal analogique (Sa) en un signal numérique (Sdi) pour une sortie par rapport à un niveau de référence ;

    un dispositif de filtrage numérique (4) tel que défini dans la revendication 1 ou 2.


     
    7. Un dispositif de détection de croisement de zéro (14) dans le dispositif de filtrage numérique (4) tel que défini dans la revendication 1 ou 2, ledit dispositif de détection de croisement de zéro étant en outre configuré pour
    générer un signal de détection de croisement de zéro lorsque le temps accumulé a atteint une valeur de référence prédéterminée.
     
    8. Un programme de filtre numérique pour réaliser informatiquement une fonction d'un dispositif de filtrage numérique tel que défini dans la revendication 1 ou 2.
     




    Drawing



























    REFERENCES CITED IN THE DESCRIPTION



    This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

    Patent documents cited in the description