JPH0315687B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a dynamic balance measuring device for measuring the dynamic unbalance (expressed by the dynamic unbalance amount and dynamic unbalance angle) of a rotating body of rotating equipment, and in particular, the present invention relates to a dynamic balance measuring device for measuring the dynamic unbalance (expressed by the dynamic unbalance amount and dynamic unbalance angle) of a rotating body of rotating equipment. The present invention relates to a measuring device for a specimen having only one surface to be corrected.

If the rotating body is unbalanced beyond the allowable limit,
Vibration and noise are generated during operation, which significantly reduces the efficiency and life of the rotating equipment, so it is necessary to measure the dynamic unbalance of the rotating body and correct the unbalanced part. Recently, there is a strong demand for high speed, high performance, and low cost for rotating equipment, so automating the motion balance measurement and subsequent unbalance correction process and improving accuracy have become issues.

By the way, calculations, storage and display in the circuit section of the measuring device of the conventional dynamic balance tester were all analog type.

However, it has become possible to relatively easily produce high-performance machines for dynamic imbalance correction at low cost using microcomputers and numerical control devices, which have developed rapidly in recent years. The device uses digital signals, and digital values are required as unbalance data to the corrector.

In addition, with the measuring instruments of conventional dynamic balance testers, the operator must carry out adjustment work called preliminary drive to set the constants of the vibration system in the measuring instruments before measurement. It is necessary to operate the potentiometers 3 to 8 and the changeover switch 9 while looking at the operator, and this adjustment is so delicate that only an expert can do it, and if the adjustment is incorrect, it will impair measurement accuracy. Furthermore, when humans read imbalances using indicators, there are problems such as errors due to individual differences in reading and reading limits due to the size of the minimum scale due to the limits of the size of the indicator.

Furthermore, as will be explained later, in order to measure dynamic unbalance, a circuit is first required that separates and calculates the amplitude and phase of the unbalance signal component that has a predetermined relationship with the dynamic unbalance from the distorted wave signal. It is. In conventional measuring instruments, a synchronous rectifier circuit and a smoothing circuit are used to convert the amplitude of only the sine wave component and cosine wave component of the unbalanced signal component in the output of the vibration detector into the corresponding DC voltage. There are many cases. Therefore, a sufficient filter effect cannot be obtained unless the time constant of the smoothing circuit is lengthened, and it takes several times the time constant for the DC voltage to stabilize, so the measurement time as a measuring instrument becomes longer. In addition, it had drawbacks such as errors due to analog calculation circuits and drift associated with DC voltage amplification.

As described above, conventional analog measuring instruments have problems such as automatic correction of unbalance, adjustment during preliminary drive, measurement accuracy measurement time, etc., and the present invention solves these problems with digital measuring instruments. This can be resolved by doing so.

Hereinafter, as an example, the case of a measuring instrument for a vertical dynamic balance tester will be explained based on FIGS. 2 to 7. Figure 2 is a panel diagram of the measuring instrument, where 10 is a number display that displays the amount of unbalance including the decimal point, and 1
1 is an unbalance angle indicator, which eliminates individual reading differences in conventional analog indicators. 1
Reference numeral 2 designates a changeover switch, which is set to Q ₂ during normal measurement and set to Q ₁ during preliminary driving, and push button switches 13, 14, 15, weight setting digital switch 16, and angle setting digital switch 17 are used. FIG. 3 shows a simplified diagram of the main body of a general vertical dynamic balance tester. When the drive motor 18 rotates, the pulleys 19, 21 and belt 20
2 rotates. The rotating main shaft 22 has a radial ball bearing 2
3. The main body of the machine is supported by the vibrating frame 24 and the spring material 25, and the rotating main shaft 22 has an adapter 27 for attaching the test specimen 26, an angle scale plate 28,
An angle reference signal plate 29 is attached. When the drive motor 18 rotates, the bearing 23 generates vibration due to the unbalance of the test object 26 and the unbalance existing in the rotating main shaft 22 and its attachments, which is detected by the vibration detector 31. As shown in FIG. 4 Sv, the output of the vibration detector 31 is a sine wave synchronized with the rotation of the rotating spindle 22, which has a predetermined relationship with the dynamic unbalance including the test object 26, the rotating spindle, and all of its appendages. becomes. In reality, the output Sv often has a waveform such as that shown in FIG. 4 Sw, in which a noise component is superimposed.

On the other hand, since the reference mark of the angle reference signal plate 29 attached to the rotating main shaft 22 is detected by the phase detector 30, the output of 30 is synchronized with a pulse once per rotation as shown in Fig. 4 Sp. And it comes out with a constant phase. Therefore, this Sp is the reference phase (0
degree).

Next, we will explain a method of calculating only the unbalance signal component in the output Sw of the vibration detector 31 (hereinafter referred to as vibration signal) containing a noise component using a microcomputer. This will be explained with reference to FIG. In FIG. 5, a flip-flop circuit 32 receives the signal Sp and outputs a signal that is inverted every cycle as shown in the signal S1 in FIG. 33 is approximately 1/1 of the period Ts of the vibration signal
The gate circuit 34 is an oscillator that oscillates pulses with a period of 000 or less, and its output is controlled by the S1 signal.
is input. In the gate circuit 34, the signal S1 is ON
Since the oscillation pulse is passed only when , the output will be like the signal S2 in the figure. The signal S2 is input to the counter 36 and counted. Therefore, assuming that the counter 36 is reset at the time To, the counter 36 counts the pulses of the oscillator 33 from the pulse output time To of the reference phase signal Sp to the output time T1 of the next pulse. Then, the oscillator 33
Since the vibration period Ts is 360 degrees, the pulse interval Q of S2 is as follows: θ≈360/M (degrees) (1). Therefore, the angle of the vibration signal at the rise of each pulse of S2 is θ°, 2θ°, 3θ°...360
°
(=Mθ°). The fourth output AD of the counter 36
As shown in the figure, the number increases sequentially as 0, 1, 2...M.

By the way, the vibration signal Sw is amplified by the preamplifier 37, and the vibration signal Sw is amplified by the preamplifier 37, and the vibration signal Sw is amplified by the A-D converter 3, which can discriminate the polarity.
8. The above-mentioned signal S2 is given as a conversion start signal to the A-D converter 38, so S
The conversion is started from the rising edge of the second pulse, and when completed, the digital conversion data DA of that voltage and the conversion completion signal pulse S3 are output. This data DA
The completion signal S3 becomes the data input and write pulse signal of the memory circuit 39, and the digital value of the oscillating voltage is stored in the memory circuit 39 at each pulse timing of S3.
write to.

Since the address input of the memory circuit 39 is connected to the output of the counter 36, the oscillating voltage at the angle θ is
S _W1 value is “1” address, angle 2θ value S _W2 is “2”
It will be remembered as a street address. That is, the value of the last vibration signal voltage S _WM in FIG. 4 is stored at address "M", and the value of the Qth S _WQ is stored at address "Q".

Now, the signal S1 is input to a microcomputer 43 (hereinafter abbreviated as microcomputer) via an interface circuit 42. The microcomputer 43 is a program memory 4 that stores programs.
According to the flow of the program No. 4, calculations to be described later are started from the fall time T1 of this signal S1, and a calculation-in-progress signal S4 is output to reset the flip-flop circuit 32. When the calculation is completed, this signal S4 is turned off. Then, since the falling differentiation circuit 45 outputs the reset pulse signal S5 at the falling edge of S4, the signal S6 causes
Counter 36 is reset. The reset state continues until T2, which is the same state as the initial state To in FIG.
If the calculation takes a long time and the calculation does not end during the period T1-T2, the calculation-in-progress signal S4 is
Since it remains ON, flip-flop circuit 3
2 remains reset and does not invert even at time T2. Therefore, the gate circuit 34 is not opened and the A/D conversion start signal S2 is not output, and the contents of the memory circuit 39 are not rewritten. If the calculation signal S4 is OFF before the next phase reference signal Sp is turned ON, the output of the flip-flop circuit 32 is inverted at the rise of this signal Sp, and becomes the initial state.

When the calculation signal S4 turns ON, the memory circuit 39
Switch from a state in which the A-D conversion data DA is written and stored at the address AD specified by the counter 36 to a state in which the stored A-D conversion data can be read out to the data bus DAT according to the address data ADD specified by the microcomputer. It is now possible to do so.
Further, the value of the output AD of the counter 36 is read out via the interface circuit 42, and the pulse interval θ of the signal S2 can be calculated using equation (1).

The calculation performed by the microcomputer 43 is to obtain the amplitude values of the sine wave component and cosine wave component of the unbalance signal component Sv, which is synchronized with the rotation of the rotating spindle, from the vibration signal Sw, using Fourier transformation.The contents are as follows. I will explain.

The microcomputer 43 first calculates the equation (1) above to determine the angular interval θ at which the AD conversion data is taken. Next, the digital values of the vibration signal voltage values S _W1 , _{S W2} . . . Calculate the sine wave component WY and cosine wave component WX of the components that are synchronized with the main shaft rotation period Ts.

W _X = 1/M {S _w1cps 〓°＋S _w2cps2 〓°＋…＋S _WMcpsM
〓°｝=1/M _M 〓 ^Q=1 S _WQcpsQ 〓° ...(2) WY=1/M _M 〓 ^Q=1 S _WQsioQ 〓° ...(3) When the noise component is large or If the period is close to the rotation period of the rotating main shaft, the noise component cannot be removed, and it is inaccurate to determine the unbalanced signal component based on the result of only one calculation. Therefore, a more accurate value can be obtained by repeating this calculation over several tens of cycles or more and calculating the average value of the answers. The operation is performed as follows.

In addition to such calculation programs, the program memory 44 described above has a storage location (RAM area) in which the microcomputer itself can freely write and read the contents. When the measurement start signal STA from the test machine body turns ON, the microcomputer senses this via the interface circuit 42, and when the S1 signal turns OFF, it first turns ON and OFF the calculation signal S4 to generate a pulse signal. Making,
Reset pulse S5 generated by the differentiating circuit 45
The counter, flip-flop, and frequency divider circuit are reset to their initial states. After that, each circuit operates as described above, and the first calculated values of equations (2) and (3) are
Calculate the values of W _X1 and W _Y1 and save the program memory.
Store the values of W _X1 and W _Y1 in the RAM area.
In the next cycle, the newly calculated values of W _X2 and W _Y2 are stored in another RAM area, and the same operation is repeated.

By repeating this calculation operation for a certain period of time (for example, 3 seconds) after the measurement start signal STA from the main body of the dynamic balance tester is turned on, Fourier transform values over several dozen cycles or more can be obtained. Next, these average values V _X and V _Y are calculated. The case of cumulative addition C times is expressed by the following equation.

V _{X =} 1/C (W _X1 +W _X2 +…+W _{XC )} = 1/C _C 〓 ^P ₌ ¹ _W C is the total number of additions, counted by the microcontroller itself,
It also requires a program to be stored in its own register or RAM area.

The values _{of V} They have the following relationship.

Next, the microcomputer 43 uses the calculated values _{of V}
The operation of calculating the angle will be described along with the operator's operating procedure.

In FIG. 5, the components numbered 10 to 17 are the same as those explained in FIG.
The status of switches 12 to 17 is detected, and the unbalance amount and angle are displayed as digital values on numerical displays 10 and 1.
Display on 1. Furthermore, for automatic correction, EX,
A digital signal output of the unbalance amount and angle called out is output from the interface circuit 42.
Note that 46 is a confirmation lamp for the operator, as will be described later, and is ON/OFF controlled by the microcomputer 43.

Now, as shown in Figure 6, check the unbalance of the test specimen 26.
Let U _W be the vector, and the unbalance due to the rotating main shaft 22 and its appendages is expressed as a vector as U→ _A , and the resultant vector is U→ _O. U→ _O is included in the output S _W of the vibration detector 31. It is proportional to the amplitude Sm of the signal component Sv synchronized with the rotating main shaft, and its direction (expressed as an unbalance angle) is represented by the phase angle ρ of Sv. If _{the composite} vector _{of V} It lags behind the composite vector U→ _O shown in FIG. 5, but its magnitude is proportional to the amplitude of the unbalanced signal component S _v , so U→
It is also proportional to the size of _O.

Therefore, if k is a positive real number, the following relational expression is established: U→ _O =kV→ _O∠δ °=k∠δ°( _VX + _jVY )...(6). (However, j ² =-1) In equation (6), k and δ are constants, but their values are still unknown.

Furthermore, there is a relationship as follows: U→ _O = U→ _W + U→ _A (7), but the value of U→ _A is also unknown, so in order to find the unbalance U→ _W of the test specimen, these It is necessary to measure an unknown constant, and this corresponds to the work called pre-driving described above.

Now, if we express U→ _A = k∠δ°(a+jb)...(8), then by substituting equations (6) and (8) into equation (7), we get U→ _W =U→ _O −U→ _A =k∠δ°{(V _X -a)+j(V _Y -b)}...(9).

Now, pre-driving is performed using any test specimen, and the operating procedure is as follows: 1. Set the selector switch 12 to the Q1 position and notify the microcomputer 43 of pre-driving.

2. After attaching the test specimen to the adapter 26, rotate the drive motor. When the vibration condition stabilizes and the measurement start signal STA is output from the machine body, the microcomputer repeats the calculation for a certain period of time as described above, and calculates _V
Calculate the value of V _Y and store the values as V _X1 and V _Y1 .

Figure 7a is a vector diagram at this time, and if the unbalance vector of the test object is U→ _W1 , then U→ ₁ = U→ _W1 + U→ _A = k∠δ° (V _X1 + jV _Y1 )...(10) Become a relationship.

When the microcomputer calculates and stores the values of V _X1 and V _Y1 , it lights up the lamp 46 to notify the operator of this.

After checking the lamp 46, the operator presses the push button switch 13 to inform the microcomputer that the state shown in FIG. 7a is present, and the lamp 42 is turned off.

3 Stop the drive motor, rotate the test specimen 26 to the adapter 27 by an arbitrary angle a degree other than 0 from the previous position, and then rotate the drive motor again. The values of V _X and V _Y calculated at this time are V _X2 ,
When V _Y2 is assumed, the vector diagram shown in Figure 7b is obtained, and U→ ₂ = U→ _W1 ∠a°+U→ _A = k∠δ° (V _X2 +jV _Y2 )...(11).

The operator confirms that the lamp 46 is lit and adjusts the angle a.
After setting the digital switch 17, the switch 14 is pressed to inform the microcomputer of the state shown in FIG. 7b.

4. Stop the drive motor, rotate the test specimen 26 by an angle a, and attach a test weight to the test specimen at an arbitrary angle γ degrees, and rotate the drive motor again. The calculated values of V _X and V _Y at this time are V _X3 ,
When V _Y3 is assumed, the vector diagram shown in Fig. 7c is obtained, and the following holds true: U→ ₃ =U→ _W1 ∠a°+U→ _A +W→ =k∠δ°(V _X3 +jV _Y3 )...(12). W→=w∠γ° (13) If the weight value is w grams in the unbalance vector due to the trial weight, then there is the following relationship.

The operator similarly confirms that the lamp 46 is lit, sets the value of w and the γ degree on the digital switches 16 and 17, presses the push button switch 15, and presses the seventh button.
Notify the microcontroller that it is in the state shown in Figure c.
Then, the microcomputer inputs the digital switch values and stores the memorized V _X1 ~V _X3 , V _Y1 ~
From equations (11), (12), and (13) using V _Y3 , δ°＝γ°−tan ^-1 V _Y3 −V _Y2 /V _X3 −V _X2 ...(15) From equations (8), (10), and (11), a=1/2{ _{−V Y1} sinα°＋ _{V X2} + ( _{V X1} cosα°＋ _{V Y1} sinα _{° −V} cosα° ₊ V _Y1 sinα°+V

Preliminary driving is completed after the above three measurements, and it becomes possible to measure the unbalance of any test variation. To perform measurements, the operator presses the selector switch 12.
Q ₂ to notify the microcontroller that it is a measurement, and the microcontroller sends this to interface circuit 4.
Upon receiving the measurement start signal STA from the main body of the tester, the values of V _X and V _Y are calculated and stored as described above, and the unbalance of the test specimen is calculated. That is, if the unbalance amount of the test specimen is U _W and the unbalance angle is W, then U → _W = U _W ∠ _W ... (18) Therefore, from equation (4), U _W = k√( _X − ) ² + ( _Y −) ²
...( ₁₉ ) A=δ°+tan ^-1 _{V Y} -b/ _V
Displayed by 1.

All of the above calculations and input/output control are performed according to the program stored in the program memory 44.

In this embodiment, the above-mentioned angles α and γ can be set variably using the digital switch 17, but in general, there are many cases where it is acceptable to fix the values of α and γ. is registered as a constant in the program memory 44 of the microcomputer, and the digital switch 17 can be omitted. The same applies to the digital switch 16.

The A/D converter used in this embodiment can resolve ±1 volt with 1 bit and 11 bit accuracy for polarity determination, and is a commercially available LSI and can be obtained at low cost. If the input value of the A-D converter exceeds 1 volt, a switch is placed in front of the A-D converter to divide the voltage and input it to the A-D converter. By adding a program to control the pressure changeover switch, it is possible to create a measuring device that can automatically measure a wide range of unbalance amounts.

On the other hand, in order to calculate the unbalance components V _X and V _Y with higher accuracy from the output Sw of the vibration detector, it is better to have less noise components at the input of the AD converter 38 . For this purpose, a bandpass filter is inserted between the preamplifier 37 and the A/D converter 38, which passes only signal components near the unbalanced signal frequency. This bandpass filter causes a phase shift of the unbalanced component, but δ in equation (6) above
Since this phase shift is included in , there is no need to change all calculation formulas.

In the embodiment shown in FIG. 5, as is clear from FIG. 4, the number of circuits is reduced by sequentially capturing and storing the vibration signals during the first period in the memory circuit and performing calculations during the next period. However, another similar memory circuit, counter, and gate circuit are provided, and when one is data, the microcomputer reads the data stored in the other memory circuit and calculates it.
If a configuration is adopted in which this is repeated alternately, unbalanced signal components can be extracted with even higher accuracy.

In the embodiment shown in FIG. 5, the values of the AD-converted vibration signal data _SW1 to _SWM are sequentially stored in the storage circuit 39 independently of the microcomputer 43. The RAM area in the program memory 44 of the microcomputer functions in the same way as a storage circuit, so by being managed by the microcomputer, it is possible to substitute microcomputer software (programs) for hardware components (circuits). That is, the memory circuit 39 as shown in FIG.
A-
The same principle can be used by adding an operation program in which the microcomputer sequentially stores the data DA of the D converter in this RAM area at the timing of the conversion end signal S3. Similarly, it is also possible to omit the differentiating circuit 45 and output the S5 signal using a microcomputer program.

FIG. 9 is a flowchart showing the procedures of various operations in the embodiment shown in FIGS. 2 to 7.

As described above, the present invention does not simply digitize the conventional motion balance measuring device using analog circuits, but uses an oscillator and a counter to determine the sampling timing in order to digitize the rotation period of the test object. Obtain the count value M of the proportional counter, convert the vibration detector signal for only one period into a digital value at the timing of the output pulse of the first oscillator, and from the result, convert the unbalanced signal component for one period to Fourier transform. As a result, stable measurement results can be obtained in a short time by obtaining an appropriate amount of information according to the rotational speed of the test object.

[Brief explanation of the drawing]

Fig. 1 is a panel diagram of a measuring instrument of a conventional dynamic balance testing machine, Fig. 2 is a panel diagram of a measuring instrument of the first embodiment of the present invention, and Fig. 3 is a simplified diagram of the main body of a vertical dynamic balance testing machine. , FIG. 4 is a waveform diagram of the output Sw of the vibration detector of the main body and the reference phase signal Sp, and a timing chart of the first embodiment, FIG. 5 is a block diagram of the first embodiment of the present invention, and FIG. 6 is a relationship diagram between the unbalance U _W of the test object, the unbalance U _A due to the rotating main shaft of the main body and its attached parts, and the composite voltage vector V _O in the circuit of the present invention. FIG. 8 is a block diagram of the second embodiment of the present invention, and FIG. 9 is a schematic flowchart of the operation of the microcomputer in the first embodiment. 10... Unbalance amount number indicator, 11... Unbalance angle number indicator, 12... Changeover switch, 13-15... Push button switch, 16, 17...
...Digital switch, 32...Flip-flop circuit, 33...Pulse oscillator, 34...Gate circuit, 36...Counter circuit, 37...Preamplifier, 38...A-D converter, 39...Memory circuit,
42...Interface circuit, 43...Microcomputer, 44...Program memory,
45...Differential circuit, 46...Confirmation lamp, EX,
OUT...External digital output signal.

Claims (1)

[Scope of Claims] 1. A mechanism section configured to apply rotational force to a test object so as not to disturb vibrations generated due to unbalance of the test object when the test object rotates; a vibration detector that converts the output into a signal; a reference phase detector that generates a pulse signal synchronized with the rotation of the test object; an oscillator that determines the timing of sampling the output of the vibration detector; a counter circuit that counts for one cycle of the output signal of the reference phase detector; and A- that converts the signal of the vibration detector into a digital value at each timing of the output pulse of the oscillator simultaneously with the counting of the counter circuit; a D converter and a first storage means for sequentially storing data for one cycle of the output signal of the reference phase detector of the A-D converter;
Assuming that the count value of the counter ^circuit is M and _the digital value stored in the storage means is _{S WQ} , _{the unbalance} signal components _W /MQ) and W _Y =1/M _M 〓 ^Q=1 S _WQ cos(360/MQ) _;
A second _{calculation means} for calculating _{the average values V} ) ² + ( _Y −) ² and φ _W = δ° + tan ⁻¹ V _Y −b/V _X −a (where a, b, k, and δ are constants); As a step, the output of the second calculation means when the test specimens are mounted in pairs at the reference positions of the mechanism section is
Store them as V _X1 and V _Y1 , and as a second step,
The output of the second calculation means with the test specimen mounted at a position rotated by α degrees with respect to the reference position.
V _X2 , V _Y2 are stored as V a second storage means for storing the outputs of the means as V _X3 and V _Y3 ; an input means for signaling the storage timing of each step; and a calculation formula for the values of the constants a, b, k, and δ. δ°＝γ°−tan ^-1 V _Y3 −V _Y2 /V _X3 −V _X2 a=1/2{−V _X1 cosα°＋V _Y1 sinα°＋V _X2 + (V _X1 cosα°＋V _Y1 sinα° _−V )cotα/2} b=1/2{−V _Y1 cosα°＋V _X1 sinα°＋V _Y2 +(−V _X1 cosα°＋V _Y1 sinα° _−V a fourth calculation means, and the U _W , φ _W
A digital motion balance measuring device comprising: an output means;