SU1325347A1 - Method of checking physical and mechanical indicators of ferromagnetic articles and device for effecting same - Google Patents

Abstract

The invention relates to the field of non-destructive testing and can be used to determine the performance of ferromagnetic products after surface and bulk hardening. The aim of the invention is to improve the control accuracy by optimizing the effect of magnetic poi (L

Description

Leu on the object of control. When implementing the method, the shape of the magnetizing signal of the magnetization unit 1 is initially selected in order to obtain maximum sensitivity to the Barkhausen jump areas, which are separated by filter 11 and integrated in the integration unit 12. According to the measurement results, the calculating unit 15 selects the form of the magnetizing current, then the form of the magnetized current is selected to optimize the signal of two similar

The invention relates to non-destructive testing and can be used to monitor the performance of ferromagnetic products after surface and bulk hardening.

The purpose of the invention is to improve the accuracy of control by optimizing the impact of fields on the object of control.

The drawing shows a device for carrying out the method.

The device contains a magnetization unit 1, a measuring transducer 2, two transducers 3 and D with reference details, a Barkhausen noise recording sensor 5, a generator 6 connected in series of calibrated signals, an analog switch 7, to the inputs of which the measuring converter 2 is connected, converters 3 and 4 for reference samples and a Barkhausen noise detection sensor 5, an analog-digital converter 8, a buffer memory block 9 and a spectrum analyzer 10 in the basis of the Haar function.

The device also contains a frequency filter 11 connected in series, connected to the output of the analog switch 7, and an integration unit 12 connected by a per unit multiplexer 13 to the first input. Which is connected to the spectrum analyzer 10 in the basis of the Haar functions, input / output interface 14 and counting -resolving unit 15, unit 16 samples with significantly different properties. After selecting the optimal magnetization current, the properties of the samples under study are monitored, the signal from which is stored in digital form in the buffer memory block 9 through the analog switch 7, decomposed into spectral components according to the Haar functions, and its physical characteristics are made according to generalized criteria technical indicators. 2 sec. f-ly, 1 ill.

synchronization, the outputs of which are connected to the control inputs of the analog switch 7 and analog-digital converter 8, and the first input is connected to the output of the integration unit 12 connected in series by the pulse normalizer 17 connected to the output of the high-pass filter 11, block 18

registering the number of Barkhausen jumps, the output of which is connected to the multiplexer 13, the polling interval control block 19, the torque function determining block 20 and the block

21 calculus num.

The device also includes a discriminator unit 22 connected in series, a distributor 23,

four time scales 24-27 connected between the distributor 23 and the inputs of the multiplexer 13 and connected sequentially reprogrammed permanent memory 28 connected to the input-output interface 14 and the torque function determination unit 20, the magnetization control controller 29, the inputs of which are connected to the water interface 14, the control and synchronization unit 16 and the interrogation interval control unit 19, and the digital-to-analog converter 30, the output of which is connected to the magnetization unit 11. The method with the device is implemented as follows.

In controlled product vozdeysviya electromagnetic field distribution and the recorded character number Barkhausen jumps, while additionally record character distribution areas Barkhausen jumps collectively peremegnichivayu- schihs domains groups magnetization in the first mode and the second mode is recorded magnetization spectral components elektrodinamichekogo transducer Haar basis function , while in the first magnetization mode the waveform of the magnetizing vani is set so that effectiveness to the maximum sensitivity distribution space character Barkhausen jumps in conjunction physico-technical indicators of responsible for the structure-rheological properties of a control object and in a second mode shape magnetization probe is set so as to provide the maximum The magnitude of the integral parameter according to the expression defining the

(PTOA) (otal), (1 etal), RArbar) Jlglg -... har. (one

where har is the spectral components of the signal of the electrodynamic converter 1, ..., that of the first reference sample and the second reference sample, significantly different in terms of the physico-technical indicators characterizing the quality of cementation or nitriding of the latter, and in the aggregate informative parameters of magnetic Barkhausen noise and spectral components of the Haar response, the electrodynamic transducer removed, judged on the aggregate of physico-technical indicators responsible for the operational Noah (residual) strength of the controlled product.

In the first magnetization mode, the cumulative and cumulative functions of various orders are used as informative components. Thus, in the first measurement mode, both the nature of the distribution of the number of Barkhausen jumps on different local time intervals of the analysis of various multiplicities, and the distribution of

Barkgausen hopping spas on specified analysis intervals. This is due to the fact that as the level of the probing magnetic field changes, there is a redistribution of not only Barkhausen jumps quantitatively, but the latter also significantly change the shape of the envelope signals (at the output of the primary converter),; directly related to the kinetics of the change in the boundary energy | - p with the displacement of the interdomain boundaries in the body of the object under control. Gradient) Eh

boundary energy change

(Where

X (the parameter of the displacement of domain boundaries) depends on the nature of the distribution of defects in the path of displacement of the domain boundaries, i.e. ultimately determined by the residual stress field in the control object.

Due to the fact that the processes of cementation and nitriding are directly conjugated to the diffusion of carbon or nitrogen in iron or other alloys, magnetic viscosity, which affects the speed of movement of the domain boundaries, influences the kinetics of the magnetization process. Thus, the processes of cementation and nitriding substantially redistribute the field of residual process stresses in the cemented or nitrided layer, the nature of the distribution of which significantly depends on the variation of the modes of the process of cementation or nitriding of the controlled products and, in turn, naturally affects the formation of the structures in the body. controlled object.

In addition, the field of residual compressive stresses in the surface layer (due to the mechanism of diffusion of carbon atoms or nitrogen in iron or other alloys) causes the redistribution of microhardness in the reinforcing layer.

The application of the criterion according to expression (1) makes it possible to further form the sounding parameters so as to ensure maximum sensitivity to the parameters being monitored, such as the depth of the hardened layer and the hardness of the core. To isolate information on the nature of the distribution of residual

technological stresses over the cross section of the hardened layer use information at local analysis intervals in the range of the existence of magnetic Barkhausen noise. Moreover, for a large sensitivity to residual technological stresses in the surface hardened layer, a measure of the deviation of magnetic noise from the normal law is used, for which information on invariant and cumulative functions that are | numerically equal

H

;

E H, 1 / N (.

N.

M,

1 / N () -, j

n m

zm;

H5 Mj lO-MjMj

Hg M s - ISMjM - 10 N + 30 M |,

where Mn is the central moments of n order, which are equal

N

M,

 1 / N 22 (.)

t

where N is the number of Barkgausen jumps.

To increase the sensitivity to the control of structural and rheological parameters that determine the quality of cementation and nitriding, information taken from local time intervals of various multiplicity, spaced along the time axis (1 / 2T), 1 / T Т er, multiples of the interval the presence of magnetic noise, which makes it possible to determine the distribution of residual technological stresses in the hardened layer and, consequently,. microhardness distribution in the latter.

The device works as follows.

Initially, a learning cycle is carried out to establish the optimal form of change in the magnetizing current in the first and second magnetization modes. In the first phase of the cycle

five

15

0

five

thirty

35

0

45

0

five

The training uses the criterion of the maximum sensitivity of the distribution of the Barkhausen jump areas to a set of monitored physico-technical indicators.

In the long-term, they interchange the witness samples with an electromagnetic field varying in time according to the established laws in the final phase of the previous (p-1) step of the iterative procedure, and again they specify the same. The value of the sounding parameters at the (n + 1) -th step of the iterative procedure. The signal taken from sensor 5 of the Barkhausen jester-registration is fed through an analog switch 7 S synchronized by the control and synchronization unit 1b to the input of the high-pass filter 11, the output of which contains Barkhausen magnetic noise information. Then the signal goes to the input of the block 22 discriminators.-The latter contains m discriminators () with a uniform separation of discrimination thresholds in amplitude and generates output pulses at the output of the latter when the discrimination levels are exceeded. These are then fed to the input of the distributor 23, which generates the distribution of pulses alternately on four converters 24-27 of the time scale, which in time can measure the time intervals between two adjacent pulses coming from the discriminator unit 23. Then, through the multiplexer 13 and the input-output interface 14, the signal arrives at block 20 for determining the moment functions, which allows MOMBHTH, i.e., functions to be calculated at local analysis intervals, multiples of the Barkhausen noise in the signal spectrum of various multiplicities, the interval of the corresponding multiplicity is set by block 19 control of the polling interval, the operation of which is synchronized by the control and synchronization unit 16. Subsequently, in block 21, cumulative calculations are calculated according to the algorithms of. the formulas cumulative and cumulative functions. This information enters the counting unit 15, which, on its basis, acts on the controller 29, which changes the shape of the magnetizing current.

Subsequently, a second training cycle is carried out to determine the optimal shape of the magnetizing current for the second magnetization mode. The operation of the device in the second phase of training is in many respects related to its work in the first phase of training.

The difference in this training phase is that the measurement information is taken alternately with the output of the measuring transducers 3 and 4 for the reference samples (in this case, the controlled parts differ significantly from each other in the aggregate of controlled physical and technical indicators of ferromagnetic products). The primary response transducers 3 and 4 utilize global and local Haar functions (of various multiplicities). At this phase of learning a multi-level iterative procedure for finding the optimal form of magnetizing current is carried out according to the criterion R (Vic "r). The signals of the measuring converters 3 and 4 alternately through the analog switch 7 are fed to the input of the analog-digital converter 8, from the output of which the information presented in digital form is fed to the input of the block 9 of the buffer memory. From the output of the buffer storage unit 9, information on the request of the control and synchronization unit 16 enters the spectrum analyzer 10 in the basis of the Haar functions, where the initial message is parameterized. The operation of the control path of the probe action on the object of control in the first and second phases is identical.

In the third phase of the training cycle, the multi-parameter models for monitoring the physical and technical parameters of ferromagnetic products (for example, the hardness of the hardened layer and core, the depth of the cemented or nitrated layer) are set explicitly. To do this, they interchange alternately on representative samples of a representative training sample by electromagnetic fields according to the first and second magnetization modes. The measurement information is taken from the transducers 3 and 4 and the Barkhausen noise detection sensor 5 and alternately (synchronously with the magnetized,

through an analog switch 7, either to the input of the 11 high-pass filter or to the input of an analog-to-digital converter 8. In the first magnetization mode, the Barkhausen magnetic noise information collection path, along with information on the distribution of

In the area of Barkgausen jumps (as in the first training cycle), information on the quantity is used. Barkgausen jumps on local intervals of analysis of various

nosti.

In the measurement mode, the device operates in the same way as in the third phase of the training cycle, the difference is that the monitored product is placed in the measuring transducer 2 and the Barkgausen sensor for detecting magnetic noise is superimposed on it, and the calculating unit 15 is not in this case implements a regression procedure, a-according to the algorithms established in the training cycle, which form multi-parameter estimates, measures (more accurately assesses) the level of

physico-technical indicators of ferromagnetic products.

Claims (2)

Invention Formula 1. A method for monitoring the physicomechanical indicators of ferromagnetic products, concludes that the controlled product is magnetized by an electromagnetic field and the nature of the distribution of the number of Barkgausen jumps is recorded, characterized in that, in order to improve the accuracy of control, the nature of the area distribution is recorded Barkgauzen jumps, choose the shape of magnetizing From the condition of the maximum sensitivity of the nature of the distribution of the Barkhausen jump areas to the monitored parameters, a second magnetization mode is introduced, the shape of the magnetizing signal at which is chosen from the maximum condition of the integral parameter R (barj , N this a) iiwr ztal1 m samples differing in combination of physicomechanical parameters. where haff is the spectral components of the signal from the reference samples, register the spectral components of the information signal in the basis of the Haar functions, and as informative components use cumulums of functions of various orders, and on the totality of the informative parameters of the Barkgauen noise and spectral components of the information signal in the basis of the Haar functions is judged on the physicomechanical parameters of the product, 2. A device for monitoring the physico-mechanical parameters of ferromagnetic products, comprising a magnetizing unit connected in series and a measuring transducer, an analog-digital converter, a spectrum analyzer based on Haar functions, a calculating unit, a generator of calibrated signals, and a control and synchronization unit whose output is connected with the control inputs of the analog-digital converter and the spectrum analyzer in the basis of the Haar functions, characterized in that, in order to improve the control accuracy, it equipped with two transducers for reference samples, a Barkhausen noise detection sensor, an analog switch connected in series to the input of which a measuring transducer is connected, two transducers for reference samples, a Barkhausen noise recording sensor and a generator of calibrated signals, a high-pass filter, an integrator, a block discriminator A. Kozoriz Zayaz 3043/58 Compiled by Yu. Glazkov. Tehred L. Serdyukova Proofreader V, But ha Circulation 776 Subscription VNIIPI USSR State Committee for inventions and discoveries 113035, Moscow, Zh-35, Raushsk nab., 4/5 Production and printing company, Uzhgorod, st. Project, 4 O five minators, four parallel-connected time-scale converters, a multiplexer, the fifth input of which is connected to the spectrum analyzer in the basis of the Haar functions, and an input-output interface connected to the calculator, a buffer memory block whose input is through an analog-to-digital converter connected to an analog switch and output to a spectrum analyzer in the basis of Haar functions, connected in series with a pulse normalizer connected to a high-pass filter, and a number of BjipK-Hausen jumps, the output of which is connected to the multiplexer, are connected in series by a polling interval control unit, connected to a control and synchronization unit, a moment function determining unit and a cumulative unit calculating and sequentially reprogrammed permanent memory, connected to a moment function determining unit and an input / output interface, a magnetization control controller, connected to the input / output interface and a digital-to-analog converter, output Which is connected to the input of the magnetization unit, the outputs of which are connected to the transducers for reference samples and the Barkhausen noise detection sensor, and the control inputs of the magnetization control controller and analog switch are connected to the control and synchronization unit. 0 five 0