WO2006045263A1

WO2006045263A1 - Method and device for determining the thickness of a test body by means of two different ultrasonic modes

Info

Publication number: WO2006045263A1
Application number: PCT/DE2005/001322
Authority: WO
Inventors: Michael Kaack; Thomas Kersting
Original assignee: Mannesmannröhren-Werke GmbH
Priority date: 2004-10-28
Filing date: 2005-07-21
Publication date: 2006-05-04
Also published as: DE102004053112B3

Abstract

The invention relates to a method for determining the thickness of a test body by means of ultrasound whereby the temperature of the test body is different from the room temperature and the running time of the ultrasonic waves reflected on the opposite surface side and scattered by the surface side of the test body are measured by taking into account the temperature dependency of the expansion speed of the ultra sonic waves in the material used for the test body by means of a reference or calibration measurement. Two ultrasonic waves types which are significantly different in relation to temperature dependency of the expansion speed are produced in the test body and both of the separately measured running times of the respective ultrasonic waves are offset in relation to the temperature dependency of the expansion speed which can be expressed as a factor or function.

Description

METHOD AND DEVICE FOR DETERMINING THE THICKNESS OF A TEST BODY BY MEANS OF TWO DIFFERENT ULTRASOUND MODES

description

The invention relates to a method for determining the thickness of a test specimen by means of ultrasound, in particular sheet metal or tube, according to the preamble of claim 1. Furthermore, the invention relates to a device for carrying out this method.

In the non-destructive wall thickness measurement by means of ultrasound, z. For example, on sheets, Uftraschalfϊmpufse in the wall are excited in the wall after the pulse-echo method starting from one side of the Bfechober- surfaces and the signals reflected from the against bausch surface again received signals. From the transit time between the emission of the pulse and the reception time and from the speed of sound in the material to be tested, the thickness of the sheet can be calculated.

Are known z. B. from EP 1 396720 A2 two ways to use the Uftraschafftechnik for non-destructive testing of workpieces. Normally, piezoelectric ultrasonic converters are used which generate ultrasonic waves which are introduced into the workpiece, frequently using a coupling-in medium, such as e.g. As water, be eingekoppeff.

In principle, with a piezoelectric ultrasonic transducer longitudinal waves, vertical and horizontal polarized Trahsversälwellen (the latter only using a viscous Koppetmittels) and Rayfeigh waves are excited.

In commercially available portable measuring devices for wall thickness measurements on steel materials, however, longitudinal waves are preferably excited, since the introduction into the test body is the easiest way to realize this.

It is also possible to use electromagnetic ultrasound transducers (EMUS) which enable the formation of ultrasonic waves as hori zontally and vertically polarized transversal veins within the test specimen by contactless energy transmission. Since the Uffraschaffanregung done here directly in the test specimen, a coupling medium is not neces sary. The basics for this purpose are described in DE 196 37 424 A1. Decisive for an exact determination of the wall thickness by means of ultrasound is, inter alia, the knowledge of the speed of sound propagation in the test specimen. As a physical regularity, it is known that the effective speed of sound has a clear dependence on the temperature of the sonicate penetrated through. It is also known that the different excited wave types have different speeds of sound as well as a different dependence on the temperature of the test piece.

In the case of wall thickness measurement, however, it is not the surface temperature which is generally well ascertainable that is decisive, but the temperature or the temperature distribution within the test specimen. In the case of inhomogeneous temperature distribution in the test specimen, the ultrasound in the respectively excited wave type passes through different temperature ranges, so that their propagation velocity also changes in each case.

Fundamental investigations by F. Richter ("Magnetic and other physical properties of X52, X60, X70 and X80 grade line pipe steels", Steel research No. 9/89, pp 417-424, Verlag Stahleisen mbH, Dusseldorf and "physical properties of steels, in particular waimfesten steels ", Thyssen research, 2nd year, 1979, Issue 2) have shown that for the investigated steels on the one hand a nearly linear temperature dependence of the speed of sound in the temperature range of 20 - 300 ⁰ C regardless of the material is present. The effective speed of sound z. B. for a X60 / X70 steel yielded about 3210 m / s at 20 ⁰ C and about 3060 m / s at 300 ⁰ C.

On the other hand, the investigations showed that at constant temperature, the effective propagation velocity of the sound present in the test specimen for different wave types or differently polarized (transversal) waves is likewise different.

Only with knowledge and consideration of the temperature or temperature distribution in the Prüf¬ body can therefore make a correct measurement of the wall thickness.

While the temperature coefficients in the case of a linear dependence of the speed of sound on the temperature at homogeneous temperature distribution for the sound waves in the steel materials to be measured can be easily determined and taken into account in the evaluation of Mess¬ devices with appropriate correction parameters, the actual temperature or the temperature distribution within the Test specimens, if any, can then only be determined with considerable metrological effort. Thus, the consideration of the temperature influence on the speed of sound and thus on the measurement result is much more difficult.

This problem is relevant in wall thickness measurement, for example, in the case of surface defects occurring on the test specimen during the production of metal sheets or pipes. These can often only be removed by regrinding, but the observance of the prescribed minimum wall thickness is to be observed.

In order to avoid any undershooting of the minimum wall thickness early, d. H. Before further processing, immediately after the grinding process, the wall thickness of the test specimen is determined and, if the wall thickness is undershot, it is sorted out of the production process.

However, depending on the intensity of the processing, heat is introduced into the workpiece as a result of the grinding process, which propagates at different speeds depending on wall thickness and material and cools down again at different speeds depending on the ambient conditions, as a result of which an inhomogeneous temperature distribution arises in the workpiece.

A wall thickness measurement immediately following the grinding process, without knowledge of the current temperature and the temperature distribution in the workpiece, inevitably leads to incorrect measurements of varying degrees. Since the heat input by grinding leads to a lower propagation speed of the sound in the test specimen and thus a too large non-existing value for the wall thickness is determined. With corresponding tolerance specifications, this can mean that sufficient wall thicknesses are still displayed, although in fact the tolerance has already been exceeded.

However, an accelerated cooling of the grinding points, especially in TM-rolled steels, is not possible because the material properties would be adversely affected by this. On the other hand, slow cooling in still air interrupts the production chain before the measurement and is thus uneconomical in terms of production technology.

The temperature of the workpiece surface to be measured with commercially available measuring instruments is not very meaningful, as it does not allow determining the relevant temperature or the temperature distribution in the test specimen.

Also relevant is this problem z. B. in Wanddickenmessungerr on be¬ in operation be¬ sensitive piping, which is warmer or colder to the ambient temperature Transport media. Here too, a temperature gradient to be taken into account for the wall thickness measurement by means of ultrasound in the tube wall arises in the tube wall.

EP 0 202497 A1 discloses a temperature-compensated ultrasonic wall thickness measurement for extruded plastics which is intended to enable a more accurate detection of the temperature distribution in the wall and thus a more exact determination of the wall thickness. In this case, conclusions are drawn about the wall thickness of the extruded material by measuring various process parameters such as cooling water temperature, extruder temperature, take-off rate and the residence times of the extruded material in air or in water via a mathematical model.

With this method, the temperature and its distribution in the wall of the Prüf¬ body can be detected only with considerable metrological effort and only selectively, which also has only a limited statement accuracy result. In addition, this method is unsuitable for wall thickness measurement on Stahlwerkstofferr.

The object of the invention is to develop a method for determining the thickness of a test specimen, in particular a sheet or tube, by means of ultrasound at a temperature deviating from the room temperature of the test specimen, which determines a thickness even with an uneven distribution of the temperature of the specimen over the thickness Test specimen without precise knowledge of the temperature, or their distribution in the specimen allows. Another object is to provide a corresponding test device.

According to the invention, this object is achieved in that, taking into account the material and its properties in the test specimen, two ultrasound wave types differing significantly in the temperature dependence of the propagation velocity are generated and the separately measured transit times of the respective ultrasound waves are related to the factor or function Expressable Temperaturabhän¬ speed of the propagation speed.

The great advantage of the method according to the invention is that the influence of the temperature distribution on the speed of propagation and thus on the measurement result can be eliminated with a very low measurement effort and simple calculation operations that can be carried out in the evaluation unit of the measuring device Any inaccuracies in the determination of the real wall thickness in ei¬ nem over the wall thickness unevenly heated specimen can be eliminated. The temperature or the temperature distribution in the test specimen no longer needs to be known for an exact wall thickness measurement, since in the method described with two ultrasound wave types generated in the test specimen and significantly different in temperature dependence, the actual wall thickness is determined by simple mathematical relationships of the physical Regularities completely independent of the temperature or their distribution is shown.

The mathematical-physical relationships for carrying out the method according to the invention in the thickness measurement on a steel sheet will be described in more detail below.

If an ultrasound pulse is coupled into a plane-parallel plate of thickness d you and if the speed of sound is known to depend on the temperature, the transit time t of the returning echo of the sound signal is v or de depending on the temperature T ,

^V = Y = V (I- ^ O

Here k is the temperature coefficient independent of the temperature and V _{0 is} the speed of sound at the temperature T = 0. The temperature zero or reference point (T = O) can arbitrarily, for. B. to room temperature, are placed.

The assumption of linearity is strictly physical - especially at very low temperatures - not given.

However, investigations on various types of steel show a very good linear reduction of the speed of sound with temperature. All deviations from the Linea¬ rity be in the range R of 2O ⁰ C to 100 ⁰ C under 1% o and in the range of 20 ° C to 200 ⁰ C under 2% o.

If different wave types are used, different propagation properties result with different temperature coefficients for different ultrasonic waves.

This circumstance is exploited to eliminate the unknown temperature at known temperature dependencies. Since both types of waves undergo the same distribution in an inhomogeneous temperature field, this method is also applicable there. The mathematical relationship for the different with an example for steel li¬-linear temperature dependence of the sound velocities affected waves is to ^¬ next, assuming an unknown but homogeneous temperature T:

^V ₂ = ^ = ^V ₀₁ - (IW

Strictly speaking, the wall thickness c / is itself temperature-dependent as well, but the influence of the temperature on the elastic properties and thus on the speed of sound is approximately 20 times greater for steel than the thermal expansion. The speed of sound changes when heated to 100 ⁰ C in steel by about 2%, wohinge¬ the wall thickness gen o changes by 1%. It remains unaffected in the further calculation, although it can easily be implemented with it.

If one considers the results under this disregard, then for a Wanddik- ke of z. B. 10 mm determine that the influence would be below 1/100 mm.

From the two equations above must now be teliminiert. Switching initially results

It lends itself here to introduce the thickness di or d ₂ as d _λ = ¹ AV ₀ J ₁ or d ₂ = Va-V ₀ 1 ₂ the

Thicknesses that would result from the running times and the sound velocities at T = O without taking into account the temperature influence.

Multiply by Ar * or, Zr ₂ and Equate returns

and finally the unknown thickness d

Considering in the following the thermal expansion of the material d = d _Q - (l + aT)

where d _{0 is} the thickness at the temperature T = O and a is the coefficient of thermal expansion, the result for the thickness d _{0 of} the test specimen is d _x d _Z -Jk ₁ -Jc ₂ ) d _n =

⁰ a- (d _ι -d ₂₎ + k _x d _x - k ₂ d ₂

The result is completely independent of the temperature and does not need to be further corrected if sound velocities at T = O (v ₀ ) and the temperature coefficients k are known. These are all material-dependent and must be determined in each case.

During the measurement, the data can advantageously either be taken from a database or determined before the measurement on the object by a suitable calibration measurement, as is also possible with conventional wall thickness measuring devices.

Because of the temperature independence of the result, this compensation method is also suitable for inhomogeneous temperature distributions. One can think of such a distribution as a stratified homogeneous distribution, which experiences both waves in the same way. The compensation then takes place with a different temperature layer by layer.

The linear dependence shown here is certainly a good approximation for many materials in the above-described temperature ranges, but considering quadratic terms, the dependence on sound velocity and temperature can be so well described for most steels that deviations only still within the measurement inaccuracies. The analytical solution is more extensive and will not be described in detail here.

Further functional relationships between the speed of sound and temperature are possible. Finally, a numerical method is also possible and easy to implement, in which the sought wall thickness is determined by eliminating the temperature directly from the data obtained from calibration measurements without explicitly assuming an analytical function. In any case, it is essential for the application of the method according to the invention that two ultrasound wave types are generated in the test specimen, which have significantly different effective propagation velocities and different temperature coefficients. This circumstance can then be used to eliminate the unknown temperature at known temperature dependencies of the sound waves.

In particular in an inhomogeneous temperature field, this method can be used advantageously if both types of waves undergo the same distribution, ie the physical dependences are comparable. It is therefore advantageous to couple the waves spatially as close as possible to one another in the test specimen or to excite them in the test specimen.

As to be generated wave types in the specimen come z. B. a horizontal and a verti¬ kal polarized transverse wave or a longitudinal and a transverse wave into consideration.

It does not matter in a soft way, the waves are generated in the specimen. The excitation of these waves in the test specimen can be carried out, for example, by means of electrodynamic ultrasonic transducers (EMUS), since in this way the excitation of the different waves in the specimen can be effected in a simple manner in a narrow effective range.

In principle, the coupling of the various ultrasonic waves can also be achieved with piezoelectric probes. The excitation of different types of waves in only one piezoelectric or EMUS probe is also conceivable.

Another major advantage of this method is that this method can be applied per se for all materials in which the wall thickness can be determined by ultrasound, such. B. also in plastics.

The only prerequisite for the successful application of the method according to the invention is that the fundamental temperature dependencies of the sound propagation of the different waves to be excited in the respective material, eg. B. on pre durchge performed reference or calibration measurements are known.

A schematically illustrated test device according to the invention is shown in the appendix. As possible for the tester according to the invention usable exciter configurations in the probe are basically electromagnetic ultrasonic transducer (EMUS) or piezoelectric vibration ungsartreger into consideration.

EMUS probes can excite transversal and / or longitudinal waves with different polarization or vibration modes in the test specimen. The waves excited with different temperature dependences in the test specimen are advantageously produced with only one EMUS test head a).

Geometrically different arranged piezoelectric exciter configurations in a test head for generating the different waves show the representations b), c), d).

Advantageously, the exciters are geometrically arranged in the test head so that the waves generated by the wall of the test body as close as possible durchschallen through as closely as possible to obtain the most accurate measurement even with inhomogeneous temperature distribution.

The different ultrasonic waves excited by the exciters from a surface side in the test specimen propagate over the wall thickness and are reflected on the opposite surface side.

The signals received by the same probes are then fed to an evaluation unit which, with the described mathematical method, performs a complete temperature compensation of the measuring signals in conjunction with previously carried out reference or calibration measurements.

The temperature-compensated measurement result is then displayed in a display unit.

Claims

claims

1. A method for determining the thickness of a test specimen, in particular ¹ sheet or tube, by means of ultrasound at a temperature deviating from the room temperature of the test specimen at which the duration of pers from a surface side of the Prüfkör¬ propagating and reflected on the opposite surface side Ul¬ traschallwellen are measured taking into account the temperature dependence of the propagation velocity of the ultrasonic waves in verwen¬ for the test body material by means of a reference or Kalibriermessung characterized in that in the test specimen two significantly differing in the temperature dependence of the propagation speed Ultraschallwel¬ lenarten be generated and the two separately measured transit times of the respective ultrasonic waves are related to the factor or function of expressible temperature dependence of the propagation velocity.

2. The method according to claim 1, characterized in that the ultrasonic waves are differently polarized transversal waves.

3. The method according to claim 1, characterized in that the ultrasonic waves are longitudinal and Transversalwel¬ len.

4. An apparatus for determining the thickness of a test specimen, in particular sheet metal or tube, with an ultrasonic waves in the test specimen generating exciter and connected to the Anre¬ ger evaluation unit for performing the method according to claim 1, characterized in that for generating the two different Ultraschall¬ wave types two correspondingly designed each electromagnetic ultrasonic transducer (EMUS) can be used.

5. A device for determining the thickness of a test specimen, in particular sheet metal or tube, with an ultrasonic waves in the test specimen generating exciter and connected to the Anre¬ ger evaluation unit for performing the method according to claim 1, characterized in that as an exciter to produce the different Ultra ¬ only one EMUS probe is used.

6. An apparatus for determining the thickness of a test specimen, in particular sheet or tube, with an ultrasonic waves in the test specimen generating exciter and connected to the exciter evaluation unit for performing the method according to claim 1, characterized in that as an exciter for generating the different ultrasonic waves two , in each case one wave-type generating piezoelectric probes ver¬ be used.

7. An apparatus for determining the thickness of a test specimen, in particular sheet or tube, with an ultrasonic waves in the test specimen generating exciter and connected to the exciter evaluation unit for performing the method according to claim 1, characterized in that as an exciter to produce the different ultrasonic waves only a piezoelectric probe is used.