Nondestructive Testing System Having Automatic Set-Up Means

This invention relates to a nondestructive testing system and more particularly to a nondestructive testing system which incorporates an automatic set-up means operable to obtain optimum results under a variety of test conditions. The system is readily operable and a minimum amount of time is required for adjustment of the instrument prior to initiating a testing operation. At the same time, the system is highly reliable, produces very accurate results and it is comparatively simple and economically manufacturable.

Nondestructive testing systems of the prior art have been capable of achieving very accurate and reliable results when properly adjusted. However, the required adjustment procedures usually have a certain degree of complexity and when an operator is unskilled or inattentive or has not taken enough time to read instruction manuals and learn proper procedures, the results obtained may be inaccurate or unreliable. As a result, defective products have sometimes been passed as being without any defect, sometimes with very expensive results. In some cases, a safety hazard may result.

This invention was evolved with the general object of overcoming disadvantages of prior nondestructive testing systems and of providing a system which can be easily and quickly set up for accurate and reliable results.

The invention is illustrated herein as applied to an eddy current type of nondestructive testing instrument and more particularly to an instrument for measuring conductivity, of a type as disclosed in U.S. Pat. No. 2,939,073 issued May 31, 1960 to Edward A. Eul, Jr. An important application for such instruments is the detection of cracks or the like which produce changes in the effective conductivity. In a conductivity measuring or flow detection instrument as disclosed in the Eul patent, the output signal obtained is desirably substantially independent of variations in spacing or dimensional relationships between a test coil and the structure being tested. To set up the instrument, the operating frequency is adjusted to a value defined by an intersection between two curves, one curve being a plot of voltage versus frequency with the test coil against a test piece and the other curve being a plot of voltage versus frequency with the coil spaced from the test piece. By operating at that frequency, the instrument is highly sensitive to changes in conductivity and thereby to cracks or other flaws while being comparatively insensitive to changes in the spacing between the test coil and the part in a test. Thus, a lift-off compensation is attained. Very accurate results can be attained with proper adjustment but such adjustment is somewhat difficult and is also time-consuming.

In a system constructed in accordance with this invention, memory means and the value of a test parameter are controlled by program means under each of a plurality of different set-up test conditions. The memory means stores the values of output signals and concurrent values of the test parameter such as the operating frequency of an eddy current instrument. After the values of a plurality of sets of signals obtained during such test conditions are stored, they are compared and an optimum parameter value is established and stored which corresponds to a predetermined correspondence between output signals during the different test conditions. The set-up means then operates to adjust the parameter to the stored optimum value for subsequent testing of test structures of unknown characteristics.

The arrangement of the invention may be utilized in conjunction with an instrument as disclosed in the Eul patent to adjust the operating frequency to obtain lift-off compensation, for subsequent measurement of conductivity and for detection of flaws which affect the measured conductivity of a part. Alternatively, the system of the invention might be used to adjust the operating frequency to obtain maximum sensitivity to changes in spacing or other changes in dimensional relationships between a test coil and a test structure, while at the same time minimizing the effects of variations in the conductivity of the test structure.

In another mode of operation, the arrangement may be used for facilitating sorting of parts of different characteristics. In this mode of operation, one of the set-up test conditions is established by testing one type of part and another test condition is established by testing a second type of part. After varying the test parameter over a certain range and storing of the output signals obtained, a value of the parameter may be established at which there is a maximum difference between the output signals obtained while testing the parts of different characteristics. Then a sorting operation may be performed in which a high degree of discrimination is obtained between the two different types of parts.

Important features of the invention relate to the adaptation of the circuitry of a nondestructive testing system for performance of the automatic set-up operations.

Additional features of the invention relate to the manner in which data is obtained and stored for use in the set-up operations. A specific feature is in an automatic balancing operation involving the incrementing and decrementing of signals produced by one or more digital-to-analog converters until a trigger level is established. The balancing operation is also used for the purpose of determining the level of a signal produced from the nondestructive testing instrumentation, such as from the output of a bridge circuit connected to an eddy current probe. By determining the change in level required to obtain a balance, data is obtained as to the level of the output signal of the nondestructive testing circuitry.

This invention contemplates other objects, features and advantages which will become more fully apparent from the following detailed description taken in conjunction with the accompanying drawings.

FIG. 1 is a plan view of an eddy current instrument which incorporates an automatic set-up means according to the invention;

FIG. 2 is a block diagram of circuitry of the instrument of FIG. 1;

FIG. 3 is a graph illustrating curves recorded during a set-up operation for lift-off compensation;

FIG. 4 is a graph illustrating curves recorded during a set-up operation prior to sorting of parts;

FIG. 5 is a graph illustrating the form of signals produced during a balancing operation;

FIGS. 6, 7, 8, 9 and 10 are flow charts illustrating the operation of the instrument

FIG. 11 shows circuitry of a voltage-controlled oscillator of the instrument; and

FIG. 12 shows circuitry of an amplifier, a trigger circuit and a meter circuit of the instrument.

Reference numeral 10 generally designates an eddy current instrument which includes automatic set-up controls and circuitry constructed in accordance with the principles of this invention.

The illustrated instrument 10 is designed for the measurement of electrical conductivity and the detection of cracks or other flaws which change the effective conductivity and also for sorting of parts having different electrical conductivities. In the operation of the instrument, it is connected through a cable 11 to a probe 12 which may be placed against the surface of a part 13 and the conductivity of the part 13 may then be read on the scale of a meter 14 having a pointer 15.

Prior to use of the instrument 10 in measuring conductivity, a lift-off compensation procedure should be followed for the purpose of rendering the measurements insensitive to variations in the spacing between the probe 12 and the surface of the part 13. Such variations in spacing may be caused by surface coatings, roughness, oxides, scales and rocking or wobbling of the probe during testing. In prior art instruments, the lift-off compensation has been performed manually and is somewhat complicated, involving repeated adjustments and readjustments of frequency and balance controls. With the automatic set-up means of this invention, however, the lift-off compensation procedure is quite simple. In particular, a control knob 16 is placed in a lift-off position as illustrated and, with a mode control knob 17 in either a "LO", "MED" or "HI" position, a scan button 18 is depressed while placing the probe 12 against the part 13, holding a shim of paper between the end of the probe 12 and the surface of the part 13. A "HOLD" signal light 20 is then energized for a short time and after it is deenergized, the operation is repeated but with the shim removed and with the end of the probe 12 against the surface of the part 13, the scan button being again depressed. The hold signal light 20 is then again energized for a short time, then becomes deenergized and is then reenergized for a short time interval to effect a blinking operation, indicating that the instrument has been properly adjusted for lift-off.

An automatic balancing operation is automatically performed as part of the lift-off compensation operation and also as part of a sorting set-up operation. However, the automatic balancing operation may be performed at any time by depression of a balance control button 21. It may also be performed manually by using a balance control knob 22 after positioning the mode switch 16 in a manual position.

The instrument may also be used for performing automatic sorting operations to discriminate between parts having substantially different characteristics which effect different responses to eddy currents. Again, a manual procedure might be performed for setting up the instrument for a sorting operation but the procedure involved is quite complicated. With the automatic set-up means of the invention, however, it is quite simple. In particular, the mode switch 16 is placed in a sort position and with the probe 12 placed against a sample part of one type, the scan button 18 is depressed and the hold light 20 is energized. After it becomes deenergized, the probe 12 is placed against another sample part of different characteristics and the scan button 18 is again depressed. The hold signal light 20 is again energized for a short time, then becomes deenergized and then it becomes energized for a short time interval, blinking to signal that the instrument is properly set-up for a sorting operation. Then, when the probe is placed against a part of unknown characteristics, the pointer 15 will read either lower or higher than a certain reference indication indicated by a line 23 on the scale of the meter 14. If it reads lower than the reference indication, it means that the part has characterstics like those of the first sample part used in performing the prior set-up operation. If it reads higher, it means that the part has characteristics like those of the second sample part used in the prior set-up operation.

As above noted, the instrument may be used in conventional fashion in a manual mode in which case, the switch 17 may be moved to a low, medium or high position for adjustment of sensitivity.

At any time, the frequency of operation may be determined by turning the control knob 17 to a "FREQ" position, and the position of the meter pointer 15 will then indicate the frequency of operation.

In many cases, it may be desirable to adjust a frequency control knob 24. For example, in the setting up of the instrument to obtain lift-off compensation, the probe 12 may be used with and without a shim to check to make sure that the same meter indication is obtained with and without the shim. If there is any difference, the frequency control 24 may be adjusted until exactly the same reference indication is obtained with and without the shim. It is noted that in either the lift-off mode or the sorting mode, the frequency control knob 24 may effect a frequency adjustment of about ten percent of the full range of the instrument while in the manual mode, adjustment of the knob 24 adjusts the frequency over the full range of the instrument.

The foregoing procedures are readily performed by any operator with minimal instruction and with minimal knowledge of the functioning of the instrument.

FIG. 2 is a block diagram of circuitry of the instrument 10. The probe 12 is connected to a bridge circuit 26 which is driven through a driver circuit 27 from a voltage controlled oscillator circuit 28. The frequency of the voltage controlled oscillator 28 is controllable by an analog signal applied from the movable contact of a potentiometer 29 which is connected between ground and a power supply terminal 30, the position of the contact of the potentiometer 29 being controlled by the knob 24. For the automatic set-up operation, the frequency of the voltage controlled oscillator 28 is controllable by an analog signal applied through line 31 from the output of a digital-to-analog converter 32. The digital-to-analog converter 32 has eight inputs connected through address circuits 33 to a central processor unit 34 of a microcomputer 35 which includes a random access memory 36 and a read only memory 37. The converter 32 also has inputs connected through lines 39 and 40 to latch circuits 41 connected to the central processing unit 34.

For control of the voltage controlled oscillator 28, control inputs thereof are connected through lines 43 and 44 to the central processor unit 34.

The central processor unit 34 is also coupled to switches which are controlled by the mode select knob 16 and the "scan" and "balance" push-butttons 18 and 21. As shown diagrammatically, the central processor unit is connected through what may be designated as scan, balance and interrupt lines to latch circuits 46 which are connected to fixed contacts of a selector switch 47 controlled by the knob 16, a switch 48 controlled by the push-button 18 and a switch 49 controlled by the push-button 21, movable contacts of the switches 47, 48 and 49 being connected to a power supply terminal 50.

In general, the operation is such that when the mode selector switch 47 is in either the lift-off or sort positions and when the switch 48 is operated by the scan button 18, the processor operates through the digital-to-analog circuit 32 to cause the frequency of the voltage controlled oscillator 28 to be changed in increments to a full range of frequencies.

At the same time, output signals from the bridge circuit 26 are effectively recorded in the random access memory 36 of the microcomputer. When two sets of signals have been recorded, the computer circuitry then operates to determine the optimum frequency of operation for the voltage controlled oscillator, then recording the optimum frequency in the random access memory 36, the oscillator 28 being then operated at that frequency.

In the preferred embodiment of the instrument as illustrated, the output signals from the bridge circuit 26 are not directly recorded but are recorded through the operation of circuitry which is also used for an automatic balancing operation as described hereinafter in connection with flow charts of the system. The circuitry includes an amplifier 52 which has an input connected to the bridge circuit and an output connected through a resistor 53 to a circuit point 54 which constitutes a summing point. Circuit point 54 is connected through a resistor 55 to the contact of a potentiometer 56 connected between ground and a power supply terminal 57. The contact of the potentiometer 56 is controlled by the balance control knob 22. In addition, the circuit point 54 is connected through resistors 59 and 60 to outputs of digital-to-analog converters 61 and 62 which have inputs connected to the address circuits 33. Control inputs of both converters 61 and 62 are connected through line 40 to latch circuits 41 and additional control inputs are connected through lines 63 and 64 to the latch circuits 41.

The converter 61 operates as a "fine" control while the converter 62 operates as a "coarse" control, the degree of control being determined by the values of the resistors 59 and 60. By way of example, the resistor 59 may have a value of 1.8 megohms while the resistor 60 may have a resistor of 18 kilohms.

The circuit point 54 is connected to the input of an amplifier circuit 66 which has three outputs connected to contacts 67, 68 and 69 engageable by a selector switch contact 70 coupled to the control knob 17 of the instrument. The contact 70 is connected to the input of an amplifier 72 which has an output connected to the input of a trigger circuit 73 which is connected through a line 74 to the central processor unit 34 and also to the input of a meter circuit 76 which is connected to the meter 14.

In the "FREQ" position of the knob 17, the selector switch contact 70 is engageable with a contact 77 which is connected to the output of a frequency-to-voltage converter circuit 78, circuit 78 having an input connected to the output of the oscillator 28.

In a "TEST" position of the knob 17, the contact 70 is engaged with a contact 79 connected through a resistor 80 to the power supply circuitry 81 which includes a battery, the condition of charge of the battery being indicated by the meter 14 in the test position of the knob 17. The power supply circuitry 81 may be of conventional construction and is not illustrated in detail.

For energization of the "HOLD" signal light 20, it is connected through a driver circuit 82 to an output of the central processor unit 34.

FIG. 3 graphically illustrates the type of response characteristics obtained with the instrument 10, showing curves 83 and 84 respectively obtained with and without a shim between the end of the probe 12 and the surface of the part 13. The curves 83 and 84 are plots of the output signal of the bridge 26 versus the frequency of operation of the voltage controlled oscillator 28. With the illustrated curves, there are five crossover points 86-90 at which the same responses are obtained with and without the shim and at which the system might be operated to obtain lift-off compensation The crossover point 90 which is closest to the highest frequency of operation is the preferred point for operation of the system and is automatically selected by the system of the invention.

In the system of the invention, the crossover point 90 is established by first operating with a shim over the entire range of frequencies of the unit, performing a scanning operation and recording in the random access memory 36 a set of signals which define the curve 83. Then, performing a scanning operation without a shim, another set of signals are recorded, thereby defining the curve 84. The microcomputer 35 then operates to determine the point 90. First, starting at the highest frequency and progressively lowering the frequency, the frequency of the first resonant peak is established in a manner as hereinafter described and then the crossover point above that frequency, i.e., the crossover point 90, is established by determining the frequency at which the output signals are equal.

FIG. 4 graphically illustrates curves 91 and 92 for explanation of the sorting operation, such curves being plots of the output of the bridge circuit 26 versus the frequency of operation of the oscillator 28. In the example, curve 91 is the curve obtained from the first sample tested, the data used in generating the curve being designated as "DATA 1". Similarly, curve 92 represents the response obtained from the second sample tested, the data being designated as "DATA 2". After recording of the data signals, the microcomputer 35 operates to determine the frequency at which there is a maximum difference between the output signals attained, the frequency being indicated by the vertical line 93 in FIG. 4. The system is then operated at the frequency indicated by line 93 and it provides a maximum degree of discrimination between samples of the type used in connection with the sorting set-up operation.

Before considering the details of the lift-off and sorting set-up operations, it will be helpful to consider the automatic balance operation which is used in recording the data for both the lift-off and sorting set-up operations. The trigger circuit 73 is used in the automatic balance operation and is set to trigger at a certain level which may be adjusted by a suitable control within the circuit. The input of the trigger circuit 73 is connected through the amplifier 72 and the selector switch contact 70 to one of the terminals 67 or 69 connected to the amplifier 66 which has an input connected to the summing point 64. The potential of the summing point 64 is determined by the sum of four signals. The first of the four signals is a signal proportional to the bridge circuit 26, applied to the resistor 53. The second signal is determined by the position of the balance potentiometer 56, applied through the resistor 55. The third signal is determined by the output of the digital-to-analog converter 61, applied through resistor 59 and the fourth signal is determined by the output of the converter 62, applied through the resistor 60.

Manual balance may be performed by adjustment of the knob 22 to adjust the position of the potentiometer 56 and to obtain a predetermined meter reading, indicated by a suitable mark on the scale of the meter 14. To obtain an automatic balance, the trigger circuit 73 is utilized. The operation may be clarified with reference to FIG. 5 and the flow chart of FIG. 10. FIG. 5 illustrates how the input signal to the trigger circuit 73 varies with time during a typical sequence in the automatic balance operation. In the operation as depicted in FIG. 5, the input signal to the trigger circuit 73, indicated by reference numeral 95, is initially below a trigger level indicated by the horizontal dotted line 96 in FIG. 5. This condition is sensed by the microcomputer 35, through the signal applied through line 74 from the output of the trigger circuit 73 and the microcomputer then operates through the "coarse" digital-to-analog converter 62 to increment the output of the converter 62 in steps, thereby increasing the input signal to the trigger circuit 73 as indicated by line 95. When the input signal to the trigger circuit 73 is incremented to a level above the trigger level, the condition is sensed by the computer 35 from the output signal applied through line 74 and the microcomputer then operates through the converter circuit 62 to decrement the level until the input to the trigger circuit 73 is below the trigger level. Then the "fine" converter 61 is brought into operation, incrementing the level of the signal in small increments until the input level is above the trigger level.

If when the balancing operation is initiated, the input level is higher than the trigger level, the "coarse" converter 62 is not initially incremented but is decremented until the level is below the trigger level, then "fine" converter 61 being then incremented until the level is above the trigger level.

Thus, in either case, whether the input level is below or above the trigger level, the final step is a "fine" increment to a level just above the trigger level. Once established, the level is maintained until another balance operation is performed.

The balancing operation is performed in response to operation of the balance push-button 21 and a similar operation is also used for measuring and recording of data in the scan operations performed in response to operation of the scan button 18 to obtain the lift-off and sorting set-up operations. In such scan operations, the frequency control converter 32 is stepped to change the frequency of operation of the oscillator 28 in steps from one end of its range to the other and preferably back to the starting end. Scanning in both directions is desirable in most applications to minimize the effect of delays between changes in frequency and the development of the corresponding output signals.

Initially and after each increment of frequency change, the balancing arrangement is used to shift the level of the output of the "coarse" converter 62 in one direction or the other until a balance is obtained. The digital data supplied to the converter 62 to obtain the balance is then recorded as an indication of the level of the output signal of the bridge circuit 26. It is noted that it is generally unnecessary and undesirable to use the "fine" converter for measurement and recording of the output signal level because of limitations on the accuracy of available converters.

The use of the balancing circuit during scanning has important advantages including the reduction in the circuitry required and the fact that there is generally only a relatively small change in output level when the frequency is stepped and hence the number of incremental or decremental changes required in the operation of the converter 62 is minimized so that the time required for the scan operation is also minimized.

FIGS. 6, 7, 8, 9 and 10 are flow charts illustrating the operation of the system. As indicated in FIG. 6, the system is energized after power is applied and then waits for an interrupt which is produced when either the scan push-button 18 or the balance push-button 21 is depressed. As indicated in FIG. 7, when an interrupt is produced, the switch status is determined and if the balance switch 49, operated by push-button 21, is then set, an automatic balance operation is performed as shown in FIG. 10 and as described hereinbefore. If the balance switch is not then set, and if the mode switch 47 which is operated by knob 16 is not in the manual position, the frequency mode is set to automatic and the hold signal is turned on in response to the set condition of the switch 48 operated by the scan button 18. The frequency mode is set to automatic through signals applied through lines 43 and 44 to the oscillator 28 and through signals applied through lines 39 and 40 to the frequency control digital-to-analog converter 32. The hold signal effects energization of the hold signal light 20 through the drive circuit 80.

Then a determination is made as to whether a "SCAN 1" flag is set. If not, a first curve is read into a "DATA 1" portion of the random access memory 36, this being accomplished by incrementing the frequency in the manner as described hereinbefore, and recording the corresponding digital signals applied to the "fine" and "coarse" converters 61 and 62 to obtain a balance. After reading the first curve, the "SCAN 1" flag is set, the hold signal is turned off and the system waits for an interrupt. Then if there has been no change in the status of the balance and mode switches 49 and 47 and if the scan switch 48 is operated by the scan button at a subsequent interrupt, a second curve is read into a "DATA 2" portion of the random access memory 36 after which the hold signal is turned off to deenergize the light 20. Then a determination is made as to whether the mode switch is in the sort position or the lift position and an operation is performed as shown in either FIG. 8 or FIG. 9.

In the sort operation as depicted in FIG. 8, a location is found at which the difference between the data recorded in the "DATA 1" and "DATA 2" portions of the memory 36 is at a maximum, the address of that location being then output to the control converter 32 for the oscillator 28 which is then operated at that frequency after which an automatic balance operation is performed.

In the lift-off operation as depicted in FIG. 9, the highest frequency positive peak in the "DATA 1" portion of the memory is determined. Such a determination is preferably accomplished by starting at the high end of the frequency range and scanning in a downward direction, while comparing the data at a pair of frequency points spaced a certain number of steps apart, for example, twenty steps, and an intermediate point halfway therebetween. If the output data for the pair of spaced points are equal and the output data for the intermediate point is a predetermined number of increments higher than that for the pair of points (eight increments for example), it indicates the existence of a positive peak at about the intermediate point. This arrangement helps to insure against indicating false peaks due to random noise signals and fluctuations.

After determining the existence and location of the highest frequency positive peak in the "DATA 1" portion of memory, a determination is made as to the location of the first crossover point at any frequency thereabove. This is preferably accomplished by incrementing the frequency, determining the first higher frequency at which the data in the "DATA 1" and "DATA 2" portions are equal, then proceeding with the frequency incrementing and determining the highest frequency thereabove at which such data are equal, and then establishing the average of such frequencies by adding the two frequencies and dividing by two. This specific procedure provides a very accurate location of the crossover point, it being noted that in the vicinity of the crossover point, the two data are very close together and they may be equal at a number of locations, especially when there are noise signals and the like. By determining and averaging the highest and lowest frequencies at which the two data are equal, an accurate measurement is obtained.

FIG. 11 shows the circuitry of the voltage controlled oscillator 28. An integrated circuit 98 which may be a type 2207 circuit, for example, has terminals connected through resistors 99 and 100 to the output of an operational amplifier 102 which may be a type TL0801. A plus input of the amplifier 102 is connected through a resistor 103 to ground and a minus input thereof is connected through a resistor 104 to the contact of potentiometer 29, through a resistor 105 to the power supply terminal 30 and through a resistor 106 to its output. Another terminal of circuit 98 is connected through a resistor 107 and line 31 to the output of the converter 32. Central terminals are connected to lines 39 and 40 which are respectively connected through resistors 109 and 110 to power supply terminal 30 and ground.

The frequency range of operation of the circuit 98 is controlled in part by the value of a capacitor 112 connected between terminals thereof and may be from about 55 KHz to 200 KHz. In the manual mode of operation, the control signals applied through lines 39 and 40 are such that the frequency is adjustable over the full range by adjustment of potentiometer 29. In the lift-off and sort modes, the frequency incremented during the scan operation from a value close to the low end of the range to a frequency close to the high end of the range through incrementing of the analog signal applied through line 31 from the digital-to-analog converter 32. By way of example, there may be on the order of 255 increments in frequency. The potentiometer 29 is effective in the lift-off mode but the range of possible adjustment is only a small fraction of the full range, on the order of 10% by way of example.

FIG. 12 shows the circuitry of the amplifier 72, trigger circuit 73 and meter circuit 76. The amplifier 72 includes an operational amplifier 114 having a plus input connected through a resistor 115 to ground and a minus input connected to switch contact 70 and also through a resistor 116 and a parallel capacitor 117 to its output. The trigger circuit 73 includes an integrated circuit 118 having one input connected through a capacitor 119 to ground and through a resistor 120 to the output of amplifier 72. A second input is connected to the contact of a threshold level adjustment potentiometer 121 which has one terminal connected to ground and its other end terminal connected through a resistor 122 to a power supply terminal 123. The output of circuit 118 is connected to line 74 and through a resistor 124 to ground.

The meter circuit 76 includes an adjustable resistor 126 for control of sensitivity, connected in series with a fixed resistor 127 to the meter 14, a pair of oppositely poled protective diodes 128 being connected across meter 14. Resistors 129-132 form a pair of divider circuits connected between a power supply terminal 133 and and the output of amplifier 72 and ground, respectively, for developing output level and reference level signals on lines 135 and 136 which are usable for control and signalling purposes.

By way of example and not by way of limitation, the components of the illustrated circuits may have values or may be of types as follows:

The probe 12, bridge circuit 26 and driver circuit 27 may be similar to probes and circuits of a Type ED-520 conductivity measuring instrument manufactured and sold by Magnaflux Corporation of Chicago, Ill., of a type as disclosed in U.S. Pat. No. 2,939,073.

The microcomputer 35 may be a Type 1802 computer manufactured by RCA. A program table for effecting the operations depicted in the flow charts of FIGS. 6-10 and described hereinbefore is as follows: ##SPC1##

It will be understood that modifications and variations may be effected without departing from the spirit and scope of the novel concepts of this invention.