JPS60313A - Ultrasonic thickness meter - Google Patents

Abstract

PURPOSE:To improve measuring accuracy by integrating clock pulses related to measurement from the first time to the prescribed number of times out of plural times and varying the zero point of a zero point calibrating circuit in proportion to the integrated value. CONSTITUTION:An R-SF/F14 outputs a square wave by regarding the decay timing of a monostable multivibrator 12 to be used as a zero point compensating circuit as a zero point and the period from the zero point to the receiving timing as the pulse width related to the plate thickness. An integration counter 19 integrates all clock pulses outputted from a conversion circuit 17 in each measuring operation. A display part 21 having decimal 4-digit structure is connected to the output side of the integration counter 17 through a decoder/driver 20. After completing the measurement of 160 times, the plate thickness corresponding to the integrated value of the integration counter 19 is displayed with 1/100mm. unit by a control circuit.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an ultrasonic thickness gauge, and more particularly to an ultrasonic thickness gauge that improves accuracy by correcting measurement deviations caused by the structure of a split probe.

Ultrasonic thickness gauges can easily and microscopically measure the thickness of plates, tubes, etc. made of metals such as steel, copper, aluminum RC, and non-metallic materials such as plastic, glass, and porcelain. This measuring device can easily measure the center of a plate, which is impossible with a meter, and is widely used in various industrial fields. There are two types of thickness measurements using ultrasonic thickness gauges: the so-called resonance method and the pulse method. The latter method takes advantage of the relationship that the round trip time of an ultrasonic pulse is proportional to the thickness of a plate made of the same material. Specifically, the probe is kept in contact with the plate and the ultrasonic wave is emitted. Calculate the time required from emitting the beam to receiving the reflected wave reflected by the other surface of the board,
The board thickness is calculated from this time and the sound velocity in the board, and is displayed digitally. This pulse method is generally popular because it has good accuracy and is easy to handle.

One of the probes used in the ultrasonic thickness gauge using the pulse method is a split type probe as shown in FIG. The division probe 1 has a transmitting transducer 4 and a receiving transducer 5 with left and right shoes 2 and 3, and a shielding plate 6 provided between them. The compression spring 7 is used to press against the plate 8, which is the member to be measured, to prevent vibrations from propagating on the surface of the plate 8 from the transmitting side to the receiving side, thereby making it possible to measure the thickness of a thin plate. However, in this split type probe 1, when the thickness of the plate 8 changes, the geometric path of the pulse changes dissimilarly as shown in I, n, and III in FIG. 1, and the plate thickness and the pulse The proportionality of path length is not perfect. For this reason, the relationship between the measured value and the actual plate thickness is as shown in the solid line in Figure 2 (if the two are adjusted to match when the plate thickness is 6 mm), and as shown in Figure 3, the relationship is as shown by the solid line in Figure 3.
．． A deviation of about 18 mm occurred. In Figures 2 and 3, the deviation changes linearly for plate thicknesses from 1 to 12.
2 to 30m is constant. This is because the range of 12 mm to 30 mm is a stable region near the focal path of the segmented probe.

If the measurement deviation due to plate thickness is left uncorrected, it will not be much of a problem when measuring with a rough accuracy of 1/10 to 17,100 m according to the user's request. If you try to make a measurement in the current state, the error will increase and the credibility of the lower digits will become less reliable, making it impractical. On the other hand, in order to perform correction, FIG.

If a table according to the correction curve of the relationship shown in Fig. 3 is stored in the memory in advance, and the actual value is read out as data from the memory using the measured value as an address, the circuit configuration becomes complicated and large. There are disadvantages in that it increases the weight and requires a huge amount of time and effort to create the program.

The present invention was made in view of the above-mentioned drawbacks of the prior art, and is an ultrasonic thickness probe that is small, lightweight, and has a simple configuration, and that improves measurement accuracy by correcting measurement deviations of split-type probes. Its purpose is to provide a

In the present invention, among multiple measurements for averaging,
By comprising a counter that integrates and counts clock pulses related to measurements from the first time to a predetermined number of times within a certain range, and a zero point pressure regulating circuit that varies the zero point of the zero point calibration circuit in proportion to the count value of this counter. , which aims to achieve the above objective.

First, the principle of the correction method of the present invention will be explained. The time interval between transmission and reception in the pulse method is the sum of the path time within the plate 8 and the path time of the shoes 2 and 3 in FIG.

For this reason, ultrasonic thickness gauges have traditionally been equipped with a zero point calibration circuit that delays the zero point by a predetermined delay time (the time it takes for the ultrasonic pulse to pass through 7 u 2 and 3) from the transmission timing.
The plate thickness is calculated at the time interval from this zero point to the reception timing. When measuring this time interval by counting clock pulses of an asynchronous oscillator, 1
If the measurement is performed only once, the measured values will vary by the size of one cycle of the clock pulse. In order to eliminate this inconvenience, an averaging method is adopted in which the plate thickness is calculated by integrating the measured values of multiple times. By the way, shifting the zero point of the zero point calibration circuit forward is equivalent to performing a ten correction, and shifting it backward is equivalent to performing a negative correction. Therefore, the present invention
Based on the plate thickness data obtained at the initial stage from the first time to the predetermined number of measurements during averaging,
The correction amount is set by varying the zero point of the zero point calibration circuit by a predetermined correction amount, and a fixed correction amount is applied to each subsequent measurement to eliminate measurement variations and ensure reliable correction. I will continue to do so.

Hereinafter, one embodiment of the present invention will be described based on FIGS. 4 to 8.

Here, in Figure 3 quoted earlier, the thickness is 1 to 6n, so a correction is necessary due to a deviation of 7. In the example, the second
As shown by the dotted line in Figure 3, the deviation is set to 0 when the plate thickness is omm, and the V is set so that the deviation is always 0. For 011 to 12 mm1, the deviation is proportional to the plate thickness. A constant 10 correction is applied after 12 m, and 160 measurements are accumulated and averaged, and the correction amount is adjusted in the first 16 measurements. The reason why the correction amount adjustment period is set to 1 for a plurality of times is to avoid variations in the measured values during this period.

FIG. 4 is a block diagram of an ultrasonic thickness gauge using a split type probe according to the present invention, and FIG. 5 is a timing chart. In FIG. 4, the split type probe 1 is pressed down closely from above to a plate material 8, which is a member to be measured, through an oil film or the like. A transmitting circuit 10 is connected to the transmitting transducer 4 of the split type probe 1, and a receiving circuit 11 is connected to the receiving transducer 5. A monostable multivibrator 12, which is an example of a zero point calibration circuit, is connected to the transmitting circuit 10. When starting Fii measurements to this monostable multivibrator 12, a start trigger is input from an external control circuit (not shown), and as soon as a trigger pulse is input, the internal state is reversed. Then, the output set to “low” in advance is “
"Hi". This “high” state is for timing C1
(Circuit constant C9 几1. After continuing for a time T determined by the output voltage U of the D/A converter 13 applied to one end of R2 and R1, the power supply voltage applied to one end of R2 and R1 continues for a time T,
Return to Low. Output voltage U of the D/A converter 13
rises step by step from U=O to U=Umax, and U increases by one step (=Δ
U), the delay time T is configured to become shorter by Δt.

Energized by the rise change of the monostable multivibrator 12 from "low" to "high", the transmitting circuit 10 generates a single pulse, and the transmitting vibrator 4 performs electro-acoustic conversion to generate an ultrasonic wave. Force the pulse to be transmitted. This ultrasonic pulse is applied to shoe 2. Receiving transducer 5 via plate material 8 and shoe 3
The signal is received by the receiver, undergoes acoustic-to-electrical conversion, and then is sent to the receiver circuit 11. This receiving circuit 1
1 is signal amplification function and automatic gain adjustment function (AGC)
, so that the amplitude of the received waveform is constant regardless of the thickness of the plate. The output side of this receiving circuit 11 is connected to the R (reset) terminal of an R-8 flip-flop (hereinafter abbreviated as 几-8F/FJ) 14. On the other hand, the 8 (set) terminal of this R-8F/F 14 is connected to the output side of the monostable multivibrator 12 via an inverting circuit 22. When the output changes from "high" to "low", it enters a set state and sets the Q terminal output to "high". On the other hand, when the reception signal is input manually from the receiving circuit 11, it enters a reset state and sets the Q terminal output to "low". do. The R-8F/F14 configured in this way has a zero point at timing 9 when the monostable multivibrator 12 falls from "high" to "low", and the period from this zero point to the reception timing is related to the plate thickness. It has the function of outputting a square wave as a pulse width.

The Q terminal of the -8F/F ratio 14 is connected to one input terminal of an AND circuit 15. An asynchronous oscillator 16 is connected to the other input terminal of this AND circuit 15, and while a square wave is input to this AND circuit 15 from H, -8F/F 14, the gate is open and the clock pulse of the oscillator 16 is is sent to a conversion circuit 17 connected to the output side. This conversion circuit 17 is based on the fact that when the speed of sound in a member changes due to a difference in the material of the member to be measured, the pulse width of the square wave output of R-8F/F14 related to the above-mentioned plate thickness changes even if the plate thickness is the same. , AND according to the sound speed set in the sound speed setting unit 18 such as a DIP switch group.
It has the function of thinning out pulses at a constant rate from the output pulse train of the circuit 15, and outputting the same number of pulses when the plate thickness is the same, regardless of the difference in material. To be more specific, when the speed of sound is high, @ estimates the plate thickness to be thinner and reduces the rate of thinning, whereas when the speed of sound is slow, it estimates the plate thickness to be thicker and increases the rate of thinning. .

An integration counter 19 serving as a plate thickness measuring circuit is connected to the output side of the conversion circuit 17. This integration counter 1
Reference numeral 9 has a function of counting the clock pulses sent out from the conversion circuit 17, and furthermore, the reference numeral 9 has a function of counting the clock pulses sent out from the conversion circuit 17, and is further configured to be able to cumulatively count all the clock pulses related to 160 measurements for averaging. has been done. Also,
The clock pulse output of the conversion circuit 17 is input to the l)/A converter 13. This D/A converter 13 includes, for example, a counter section (not shown) that counts clock pulses output from the conversion circuit 17, and a ladder circuit (not shown) connected to the bit terminal of each digit of this counter section. Thus, each time a clock pulse is input from the outside, the count value of the counter increases by 1, and a voltage that increases by 1 step (=ΔU) in proportion to this count value is output from the ladder circuit. There is.

This D/A converter 13 performs conversion from the first to the 16th measurement out of the 160 repeated measurements, or if the count value of the counter reaches "n" before the 16th measurement, until it reaches rllJ. While integrating all the clock pulses sent from the circuit 17, an analog voltage corresponding to the integrated value is output, and when the 16th measurement is completed or the counted value becomes "n" before the 16th measurement, the counter section After being held, the output voltage value U (16) or Un is fixed at that time.

The reason why the integration is stopped when the count value of the counter reaches "n" is to keep the correction amount constant after the plate thickness reaches 12 mm in FIG. 3, and the value of n will be described later.

The output voltage U of this D/A converter 13 is applied to one end of R, 2 forming the timing CR circuit of the monostable multivibrator 12, and the capacitor C is charged depending on the magnitude of this voltage U. The current changes in magnitude. Since U increases each time a measurement is completed, the delay time T until the voltage at the left terminal 12A reaches a predetermined threshold due to charging after the capacitor C is discharged by being energized by the trigger pulse. becomes shorter as time passes. The delay time reduction amount τ (correction amount K) is set to vary in proportion to U, with the voltage U of the D/A converter 13 being O as a reference. When one measurement is completed, the zero point of the zero point calibration circuit moves forward, so the next measurement will be 1(-8F).
/The pulse width of the square wave output by F14 is long < tX,
Therefore, the correction will be made in the direction of increasing the plate thickness. This operation is performed from the first time to the 16th time or before the 16th time.
When the count value of the counter section of the converter 13 reaches [nJ, the same procedure is performed sequentially up to that number of times, and it is set as a correction amount adjustment period. When the 16 measurements are completed or when "n" is reached, a constant correction amount K (1
6) or Kn, and the measurement is repeated up to the 160th time. Changes in the amount of correction per unit thickness at the end of each measurement from the first to the 16th (0 to 12 m)
FIGS. 6 and 7 show how the correction amount changes using the g-mass and the plate thickness as parameters, and FIG. 8 shows the relationship between the plate thickness and the correction amount at the end of the 16th test. The constant of the timing CR circuit of the monostable multivibrator 12, the applied voltage of R1 + the step voltage ΔU of the V1 A/D converter 13
is set so that the diagram in FIG. 8 and the dotted line in FIG. 3 match.

In each measurement operation, all of the clock pulses output from the conversion circuit 17 are integrated by an integration counter 19. A 4-digit decimal display section 21 is connected to the output side of the integration counter 19 via a decoder/driver 20, and when the 160 measurements are completed, the integration counter 19 is activated by a control circuit (not shown). The plate thickness corresponding to the integrated value of the counter 19 is displayed in units of 1/100 m.

Next, the overall operation of the above embodiment will be explained.

Here, an averaging method is used to measure the plate thickness by accumulating the measured values of 160 times, but when the accumulated value of the 160 times on the integration counter 19 is 100, the display section shows 1.
．． It is assumed that it is configured so that it is displayed as 00 yra. At this time, the count value n held by the counter section of the D/A converter 13 before performing 16 measurements is the integrated value of 16 measurements of the plate thickness of 12.00 m, and is 60.

First, when the sound speed corresponding to the material of the plate material 8 is set in the sound speed setting section 18 and the device is operated, the R-8F/
F14 is reset, and the integration counter 19 and the D/A converter 130 counter section are cleared.

Therefore, the output of the D/A converter 13 becomes O.

In this state, the control circuit outputs the first trigger pulse to the monostable multivibrator 12 (see FIG. 5). When 12 monostable multivibrator trigger pulses are input, the output is inverted, and at this rising edge, the transmitting circuit 10 generates a single pulse, causing the transmitting vibrator 4 to transmit an ultrasonic pulse. Also, the initial delay time T.

When the monostable multivibrator 12 rotates again after the period has elapsed, R-8F/F14 is set with this timing as the zero point. When the ultrasonic pulse is reflected by the plate 8 and received by the receiving transducer 51C, it passes through the receiving circuit 11 and is transmitted to the 几-8F/F.
Is the R-8F/PI sent to l4? Force reset. While this 几-8F/F14 is in the set state, AND
A clock pulse is output from the circuit 15. These clock pulses are thinned out in accordance with the speed of sound in a conversion circuit 17, and then counted by an integration counter 19, and also counted by a D/A converter 13, and a corresponding analog voltage U(1) is output. In response to the change in the output of the D/A converter 13, the delay time T of the monostable multivibrator 12 becomes shorter than T o by τ(1). In other words, in the next measurement, a correction K(1) is made by the amount of τ(1).

Next, the control circuit outputs a second trigger pulse to the monostable multivibrator 12. In response to this trigger pulse, the monostable multivibrator 12 is reversed, and the pulse circulates through the transmitting circuit 10, the transmitting vibrator 4, the plate 8, the receiving vibrator 5, and the receiving circuit 11 in the same manner as described above. R-8F/ni' 1
4 is the result of the first measurement, τ is higher in the second measurement than in the first measurement.
(IJ is set earlier, and it takes a longer time to be reset by the output of the receiving circuit 11. Therefore, the conversion circuit 17 outputs clock pulses whose number is increased by the correction amount K(1). This clock pulse is counted by integrating the first count value in the integration counter 19, and is also integrated and counted by the 1)/A converter 13. Therefore, the output of l)/A converter 13 is U
(U(2) is larger than twice υ, and the delay time T is also shortened by rl+, the difference τ(2), and the correction amount K(2) proportional to τ(2) for the third measurement. If the same operation is repeated 16 times, the correction amount K will increase according to Fig. 6 when the plate thickness is θ to 12 times, and the correction amount adjustment period will be 16 times. When finished, a correction amount K (16) that changes in proportion to the plate thickness is obtained (7th
(see figure). This correction amount K (16) is calculated by the dotted line A in FIG.
- Corresponds to H. Further, when the plate thickness is 12 mm or more and the count value of the counter section of the D/A converter 13 reaches 120 before completing 16 measurements, the counter section is held and the count value of the D/A converter 13 is The output is Uuo (=
120XΔU), and therefore the amount of shortening for subsequent measurements is constant, and the amount of h correction depending on the plate thickness is +t.

(See Figure 7). This correction amount is the dotted line B- in Figure 3.
Corresponds to C.

In this way, when the correction amount adjustment period ends, the counter section of the A/D converter 13 is held and fixed at the desired correction amount shown by the dotted line in FIG. 3, and the same measurement operation is repeated up to 160 times. returned.

The clock pulses associated with each plate thickness measurement from the first time, including the correction amount, are successively integrated in the integration counter 19. When the 160 measurements are completed, the integration counter 19
The integrated value is converted into a BCD code by the decoder/driver 20 and displayed on the display unit 21 as the corrected plate thickness after averaging.

According to this embodiment, during the correction amount adjustment period, the D/A converter sequentially integrates the φ) pulses related to each plate thickness measurement value and outputs the corresponding voltage to shorten the delay time of the monostable multivibrator. Since the correction is made according to the plate thickness, the correction according to the plate thickness can be automatically performed by 11Cri with an extremely simple configuration.
In addition, the final correction amount can be determined at the initial stage of 1/IOK among multiple measurements performed for averaging, and the correction can be performed successively from the first to the 16th measurement. Therefore, the existence of the correction amount adjustment period has almost no effect on measurement accuracy, and it becomes possible to bring out the effect of improving and stabilizing measurement accuracy by averaging. In addition, since the correction amount adjustment period is included in the main measurement that performs averaging, the overall measurement time is shorter than when the correction amount adjustment period is performed as a pre-measurement (preliminary measurement) before the main measurement. It is possible to do this.

In the above embodiment, the delay time of the monostable multivibrator was set by the time constant of the timing RC circuit, but it can also be set by counting the oscillation period of the clock oscillator by a predetermined number. In this case, the count number may be reduced by 7 for correction.

As explained above, according to the present invention, since the zero point of the zero point calibration circuit is varied according to the plate thickness data and the correction amount is set, it is extremely easy to do so without requiring extensive arithmetic processing.・Easy configuration and automatic correction according to plate thickness. In addition, since the correction amount is adjusted at the initial stage of lasing, the total measurement time does not become long, and since a fixed correction amount is applied after the adjustment period has elapsed, averaging can be performed without causing variations in measured values. It is possible to carry out corrections that fully demonstrate the effects of.

[Brief explanation of drawings]

Fig. 1 is a schematic diagram showing the measurement state of the split type probe, and Fig. 2 is a diagram showing the relationship between the measured value and the actual value when measuring plate thickness using the split type probe. Fig. 3 is a diagram showing the relationship between the measured values and deviations in Fig. 2, Fig. 4 is a block diagram of an ultrasonic thickness gauge according to an embodiment of the present invention, Fig. WJ5 is a timing chart, and Fig. 6 7 is an explanatory diagram of the operation of the ultrasonic thickness needle shown in FIG. 4, and FIG. 8 is a diagram showing the relationship between the plate thickness and the correction amount in the ultrasonic thickness gauge shown in FIG. 4. DESCRIPTION OF SYMBOLS 1... Split type probe, 8... Plate material, 12... Monostable multivibrator, 13... D/A converter,
16...Asynchronous oscillator, 19... Integration counter.

Claims (1)

[Claims] (1) A split-type probe that transmits and receives ultrasonic waves through the member to be measured, a zero point calibration circuit that delays the zero point by a predetermined delay time from the wave sending timing, and a clock pulse that outputs a clock pulse from this zero point to the wave receiving timing. Equipped with a measuring circuit for counting,
In an ultrasonic thickness meter that measures plate thickness using the integrated value of multiple measurements, a counter that integrates and counts clock pulses from the first to a predetermined number of measurements within a certain range. , and a zero point adjustment circuit that varies the zero point of the zero point calibration circuit in proportion to the counted value of the counter.