Process for determining the edge layer condition of objects

The invention relates to a method for detecting the boundary layer state of bodies. The invention further relates to the use of this method.

In addition to carburizing, nitriding and nitrocarburizing are currently the most important and most common methods of thermochemical surface treatment. These are used to adjust the properties of components and tools with regard to wear, fatigue and corrosion resistance in accordance with the requirements placed on them . Despite the great importance of the processes mentioned, it has not yet been possible to control and control nitriding and nitrocarburizing processes with sufficient accuracy.

Control devices for nitriding or nitrocarburizing in a gas stream are known from practice, which are limited to the detection of the atmospheric conditions or the atmospheric state in the treatment recipient during the treatment. There are essentially NH₃Analyzers, but also H₂-, CO-, CO₂-, CH₄-Analysers used to record the atmospheric condition. Mass spectrometers are also occasionally used to record the composition of the atmosphere.

Using the nitriding index determined from these measurements, the desired composition of the atmosphere is set on the basis of an experimentally determined teaching diagram by influencing the gas flows in the treatment recipient. However, there is no causal relationship between the composition of the atmosphere and the result of the surface treatment, since other relevant influencing factors, such as furnace condition, batch size, material and material surface condition, are not taken into account.

From the hardening technical report, Volume 41, No. 4, July / August 1986, R. Konrad et al .: "Non-destructive determination of heat-influencing outer layers and their thickness", pp. 213 to 218, it is known to use a thermochemical outer layer Edge treatment to be checked non-destructively.

GB-A-2184553 discloses an apparatus for determining phase change in thermoformed metals and alloys.

From EP-A-100 009 it is known to determine the hardness profile of a surface-hardened material sample, for example, non-destructively. Magnetic measuring methods are used for this after surface hardening or other heat treatment. The same is evident from Patent Abstracts of Japan, Volume 9, No. 37 (P-335) (1760) of February 16, 1985 on JP-A-59178356 of Nippon Denshi KK of October 9, 1984.

The cited prior art has the disadvantage that it offers no possibility of detecting the surface condition of a workpiece during the surface treatment and then influencing the surface treatment.

The invention is therefore based on the object of offering a method for "in situ" detection and influencing the boundary layer state of bodies which are subjected to a thermochemical boundary layer treatment.

A method for solving this problem has the measures of claim 1. The method according to the invention offers the advantage that, in contrast to the known methods, it is not a cause of the surface layer treatment, such as the atmospheric condition, that is checked, but the result of the surface layer treatment, namely the current surface layer state, is recorded. By monitoring the actual state of the boundary layer, it is possible to regulate the boundary layer treatment, in particular the atmospheric state, in order to achieve a predetermined target state of the boundary layer. Depending on the setting of the (test) frequency, the measurement can be limited to a certain boundary layer area.

With a thermochemical (boundary layer) treatment, the measurement of the electromagnetic variable (s) takes place depending on the material of the body and the particular type of treatment at one or more (test) frequencies. This method is particularly suitable for surface treatment of metallic materials by nitriding or nitrocarburizing.

The complex electromagnetic immunity is measured, which provides information about the electrical and magnetic state of the material. To limit the effort involved in carrying out the method, the measurement is carried out on a material sample of the same material, material state and material surface layer condition as the rest of the batch subjected to the surface layer treatment. The method also offers the possibility of measuring the immunity of a combination of the body (sample) with electrotechnical components.

Based on calibration curves, the measured immunity values ​​allow direct conclusions to be drawn about the parameters that characterize the surface layer condition, such as the depth of hardening, the depth of diffusion and the thickness of the connection layer.

The temperature-dependent complex immunity can be determined by cooling or heating the body (sample). This leads to the determination of the phase composition in the connection layer and in the (underlying) diffusion layer of the measured boundary layer area by determining the phase transition temperature (eg Curie temperatures).

The size of the change in immunity at the Curie point of a phase is a quantitative measure for the phase in question. In the case of composite phases with concentration-dependent Curie points, in particular in the ferromagnetic phases Fe₄N, Fe₄ (C, N), Fe_2-3 N, Fe_2-3 (C, N), the measurement of the change in immunity is used to determine the concentration of carbon and nitrogen. The thickness of the connection layer composed of ferromagnetic phases is determined from the sum of the changes in immunity in the Curie points.

The method according to the invention enables the boundary layer treatment to be controlled in order to achieve certain specifications, such as connection layer thickness, hardening depth, connection layer composition and nitriding hardness depth. In connection with the atmospheric regulation in the treatment recipient, a fully automatic control and regulation of the boundary layer treatment can be carried out to achieve a predetermined boundary layer condition.

The invention is explained below with reference to the drawings. In this show:<dl id="dl0001"><dt>Fig. 1</dt><dd>a typical resistance curve at a measuring frequency of 10 kHz,</dd><dt>Fig. 2</dt><dd>the relationship between phase composition or connection layer thickness and the relative resistance at a measuring frequency of 10 kHz,</dd><dt>Fig. 3</dt><dd>a cooling curve showing the temperature-dependent inductance,</dd><dt>Fig. 4</dt><dd>another cooling curve with the dependencies according to FIG. 3,</dd><dt>Fig. 5</dt><dd>the dependence of the Curie point of the Fe_2-3N phase from nitrogen concentration, and</dd><dt>Fig. 6</dt><dd>the dependence of the Curie point of the Fe_2-3CN phase from nitrogen and carbon concentration.</dd></dl>

The surface condition of bodies (material samples) is recorded by measuring electromagnetic variables during a surface treatment, such as nitriding or nitrocarburizing, in the treatment recipient. According to the invention, the measurement is carried out “in situ”.

This is measured with an alternating current signal (AC). In this way, the "skin effect" known from measurement technology can be exploited, which manifests itself in the fact that, with increasing frequency, an electromagnetic current is limited to a marginal area that becomes smaller. Thus, by setting an appropriate frequency, the edge area to be measured can be set variably, so that electrical and magnetic properties of samples can be determined selectively down to very thin edge layer areas (µm). The skin effect also has the consequence that the measurement signal is increased by reducing a current-carrying sample cross section, and thus a significantly greater measurement sensitivity is achieved than in the case of the same-voltage measurement.

The combination of different electrical conductors is described mathematically as a network system. If so-called bipoles such as resistors R, capacitances C and inductors L are used, but no current-generating conductors, one speaks of linear, passive network elements.

If an AC voltage U is applied to a linear, passive network system, a current I flows. In the complex notation, both quantities can be represented as follows:<maths id="math0001" num=""><math display="block"><mrow><msub><mrow><mtext>U = U</mtext></mrow><mrow><mtext>O</mtext></mrow></msub><msup><mrow><mtext> · E</mtext></mrow><mrow><mtext>jwt</mtext></mrow></msup></mrow></math><img file="EP0324809B1_D0001.tif" /></maths><maths id="math0002" num=""><math display="block"><mrow><msub><mrow><mtext>I = I</mtext></mrow><mrow><mtext>O</mtext></mrow></msub><msup><mrow><mtext> · E</mtext></mrow><mrow><mtext>j (wt-θ)</mtext></mrow></msup></mrow></math><img file="EP0324809B1_D0002.tif" /></maths><dl id="dl0002" compact="compact"><dt>U:</dt><dd>complex AC voltage</dd><dt>U_O:</dt><dd>RMS voltage</dd><dt>I:</dt><dd>complex alternating current</dd><dt>I._O:</dt><dd>Effective current</dd><dt>w:</dt><dd>Angular frequency (w = 2πf)</dd><dt>θ:</dt><dd>Phase angle</dd><dt>t:</dt><dd>time</dd></dl>

From Ohm's law, the complex alternating current resistance, the impedance Z, is calculated as a function of the frequency as the quotient of the two quantities:<maths id="math0003" num=""><math display="block"><mrow><mtext>Z (w) = U / I = </mtext><mfrac><mrow><msub><mrow><mtext>U</mtext></mrow><mrow><mtext>O</mtext></mrow></msub><msup><mrow><mtext> · E</mtext></mrow><mrow><mtext>jwt</mtext></mrow></msup></mrow><mrow><msub><mrow><mtext>I.</mtext></mrow><mrow><mtext>O</mtext></mrow></msub><mtext> · E</mtext><msup><mrow><mtext>​</mtext></mrow><mrow><mtext>j</mtext><mfenced open="(" close=")"><mrow><mtext>wt + θ</mtext></mrow></mfenced></mrow></msup></mrow></mfrac></mrow></math><img file="EP0324809B1_D0003.tif" /></maths><maths id="math0004" num=""><math display="block"><mrow><msup><mrow><mtext>Z (W) = / Z / · e</mtext></mrow><mrow><mtext>jθ</mtext></mrow></msup></mrow></math><img file="EP0324809B1_D0004.tif" /></maths> Analogue to the impedance, the AC conductance, the admittance Y, is defined. The following applies:<maths id="math0005" num=""><math display="block"><mrow><mtext>Z (w) = 1 / Y (w)</mtext></mrow></math><img file="EP0324809B1_D0005.tif" /></maths>

Impedance and admittance are summarized under the generic term immitance.

In general, the voltage, the current and the phase angle are measured and all further subsequent variables are calculated from these variables.

The measuring device used to carry out the method essentially consists of an immunity tester, commonly referred to as an LCR current-voltage measuring device, a sensor with an interchangeable measuring sample and a (shielded) electrical line between the sensor in the treatment recipient and the immunity measuring device with appropriate adaptation to the sensor.

The sensor consists of a temperature-stable, chemically stable, electrical feedthrough in the treatment recipient with a temperature-stable, chemically stable electrical insulation of the measuring leads (Al₂ O₃, Mica), whereby the measuring leads are provided with a shield (Ti (C, N) coated sheathing or contacts) to avoid measurement errors, such as the drifting of the measuring signal, and an electromagnetic (inductive) coupling of the leads to the sample .

For nitriding (32 h) a material sample made from 31CrMoV9 at 520 ° C and a nitriding index of k_n = 0.4 when the method is carried out with the measuring device described in the introduction, the values ​​explained in more detail below with reference to FIGS. 1 to 6 result.

1 shows a typical resistance curve at a measuring frequency of 10 kHz, determined from the immunity measurement, during the entire nitriding process, including heating and cooling. Here you can see the typical exponential course of the measured values ​​depending on the nitriding time, as is to be expected in all diffusion-controlled processes.

2 shows three measurements on three completely identical sample starting states during identical nitriding treatments. As accompanying studies showed, samples 11 and 12 are identical boundary layer states both in terms of the layer composition and the thickness of the compound and diffusion layer. In contrast, sample 13 shows how even small differences in the phase composition and thickness of the connecting layer are reflected in the measurement signal. It can be seen how sensitively the measurement signal reflects the true state of the nitriding treatment.

3 shows a cooling measurement. The course of the inductance of the measurement sample from the immunity measurement is shown as a function of the temperature. The measuring frequency is 1 MKz. This cooling curve was recorded with the sample 11 at the end of the nitriding treatment in FIG. 2. It is identical to the course of the test sample 12. At 482 ° C you can clearly see the Curie point of the Fe₄N phase because the inductance suddenly rises despite the falling temperature or has at least one breakpoint in the case of small portions of this phase. The curve is offset in parallel. The size of the parallel shift is proportional to the phase component in the measured volume and can therefore be used for quantitative phase determination by means of a calibration measurement.

4 shows the concentration determination by determining the Curie temperature (s). The cooling curve profile shown was recorded at the end of the nitriding treatment of sample 13 (shown in FIG. 2). The measuring frequency is 1 MHz. Here you can see another Curie point at 345 ° C besides that at 482 ° C.

5 is the dependence of the Curie point of Fe₂-₃N phase represented by the nitrogen concentration. Fig. 6 shows the dependence of the Curie point of the Fe_2-3CN phase from nitrogen + carbon concentration. The assignment here results in an additional ε-iron carbonitride phase with a concentration of 5.5% by weight.