AC Coercimeters for Quality Control of Products Made of Ferromagnetic Materials

B. B. Vinokurov, V. P. Dolgopolov, I. G. Leshchenko

(Presented by the scientific and technical seminar of the Faculty of Automation and Computer Engineering)

In the practice of quality control of products by electromagnetic methods, DC coercimeters have become widely used; their operating principle is based on measuring the coercive force of a product that has been preliminarily magnetized to a specified value. There is a large number of various measuring circuits and designs of magnetizing devices. When unambiguous relationships exist between the tested characteristic of a product and the coercive force, the problem of quality control of a ferromagnetic product or material is solved relatively simply. In the case of large parts, when the method of local magnetization is used, and especially when it is necessary to inspect the surface layer of products, DC coercimeters satisfy practical requirements less effectively.

Along with the indicated instruments for quality control of products, new types of instruments—AC coercimeters being developed at the Tomsk Polytechnic Institute—may be used. The operating principle of AC coercimeters has been published in [1] and consists in the fact that the reversal moments of products of identical dimensions in an alternating magnetic field of specified intensity are proportional to their coercive forces. Ferresonance sensors are used as zero-field indicators; they produce short emf pulses at the moment of reversal of the product. Since in practice parts are most often sorted into acceptable and rejected ones under various, sometimes nonconstant, operating conditions and with fluctuations in the voltage of the AC power source, instruments based on simultaneous comparison of the tested product with two reference specimens having permissible limiting characteristics are of greatest interest for quality control of products, or only for their rejection. AC coercimeters are of no less interest for determining the absolute value of the coercive force, for quality control of a product at a specified thickness of the surface layer, for measuring the thickness of a nonmagnetic layer on the surface of a ferromagnet, etc.

A characteristic feature of all AC coercimeters, in contrast to DC coercimeters, is that sorting of products can be performed at magnetizing currents of various amplitudes and, most importantly, at the desired thickness of the surface layer of the product being inspected.

In addition to feed-through magnetizing devices [1] (Fig. 1, a), at present surface-mounted, slotted-type devices and feed-through devices with an additional magnetic circuit have been developed and investigated (Fig. 1, b, c, d).

Each of the magnetizing devices is used depending on the shape and dimensions of the products being inspected: a) for cylindrical products; b) for inspecting a surface area; c) for thin plates and parts of the spring type, for example manometric springs; d) for long tubes and rods, etc.

Fig. 1

Figure 2, b shows the design of a zero-field sensor in the form of a permalloy probe highly sensitive to weak fields, consisting of a wire of annealed permalloy 25–100 μm in diameter and 8–10 mm long, with a signal winding of 100–500 turns of wire 0.02–0.07 mm in diameter.

In ac coercimeters, a certain difficulty is presented by the fabrication and adjustment of zero-field sensors, by circuits for analyzing the sequence in which pulses from the two reference products and the product under investigation arrive, and by signaling circuits.

The purpose of preliminary adjustment of the fabricated zero-field sensors and magnetizing devices is to obtain, without the specimens being inspected, exact coincidence of the instants at which the magnetic flux passes through zero; when the magnetizing devices are switched on successively, these instants may fail to coincide in time for various reasons (unequal geometrical dimensions of the products, properties of the material of the magnetizing devices, the influence of the width of the hysteresis loop of the permalloy zero-field sensors, etc.).

Checking and adjustment are obtained most simply and quite effectively by successively comparing with one another the emf pulses from the zero-field sensors at several values of the magnetizing current and within a specified frequency range. When two sensors are connected in opposition, without products being inspected (or with products having deliberately identical characteristics), the screen of a high-sensitivity oscilloscope shows the resultant curve from these emf pulses.

Figure 3 presents possible and most characteristic cases of obtaining resultant pulses, on the assumption that the emf pulses are symmetrical, although from each sensor they may in practice differ in shape and amplitude.

In the first three cases (a, b, and c) there is a mismatch of the remagnetization instants and, consequently, correction of the field is required. In the fourth case the emf pulses (a symmetrical double-humped emf) coincide exactly in phase, i.e., the instants at which the magnetic flux passes through zero are equal. The resolving power of determining the phase shift by the indicated method at a frequency of 50 Hz is of the order of tenths of a microsecond, which corresponds to a difference in coercive forces for structural and electrical steels by one or two orders

Fig. 2.
1 — dielectric bushings; 2 — sensor frame; 3 — sensor winding;
4 — quartz-glass tube; 5 — permalloy core.

Visible labels in the figure: oscilloscope; sample M; sample K; sample T; \(L_1\), \(L_2\), \(L_3\); \(D_m\), \(D_k\), \(D_t\); \(A\); \(R\); \(10\); \(5\); \(\varnothing 3\); a); b).

Fig. 3

Subfigures: a), b), c), d). Visible curve labels: \(e_1\), \(e_2\).

is somewhat less than is practically necessary, proceeding from the tolerances for rejecting parts according to particular characteristics that have a more or less noticeable unambiguous relation to the coercive force.

For quality control of parts made of gray cast iron and other structural steels by hardness, the KPT-2 alternating-current coercimeter with replaceable magnetizing devices has been developed. A description of the instrument and of the attachable magnetizing devices is given below.

Figure 2,a shows the circuit of the instrument with an electronic oscilloscope. It consists of three identical magnetizing devices, the magnetic cores of which are made of a magnetically soft material. The overall dimensions of the magnetizing devices are \(21 \times 26 \times 6\ \mathrm{mm}^3\), the area of the pole piece is \(6 \times 6\ \mathrm{mm}^2\), and the number of turns is \(W = 250\). The magnetizing coils are designed for a maximum current of \(0.6\ \mathrm{A}\). The design of the zero-field sensor is shown in Fig. 2,b.

Figure 4,a,b presents the block diagram of the alternating-current coercimeter for hardness rejection and the timing diagrams. \(e_m\), \(e_x\)

a)

b)

Fig. 4

and \(e_t\)—pulses coming respectively from the sensors \(D_m\), \(D_x\), and \(D_t\)—arrive through three channels consisting of three identical pulse amplifiers. The pulse \(e_x\), amplified by amplifier \(U3\), starts a waiting

a multivibrator, which produces a rectangular pulse of a duration deliberately exceeding the time interval between the pulses \(e_m\) and \(e_T\). This voltage pulse prepares both coincidence circuits, to the second inputs of which the signals \(e_m\) and \(e_T\) are applied from amplifiers \(Y_1\) and \(Y_2\).

Depending on the combination of pulse “triplets,” operation of the circuit will differ:

1. The pulse \(e_x\) is located in time between the pulses \(e_m\) and \(e_T\). This case corresponds to an acceptable part. From Fig. 4, b it is seen that only the second coincidence circuit is triggered, and its signal switches trigger \(T_2\) into the second stable state. In the trigger circuit there is a relay, which is energized at the moment the trigger switches over and, by its contacts, performs the corresponding commutation. The display “Part acceptable” lights up.

2. The pulse \(e_x\) arrives in time earlier than the pulses \(e_m\) and \(e_T\). This corresponds to a rejected part due to insufficient hardness. The pulses \(e_m\) and \(e_T\) pass through both coincidence circuits and switch both triggers. As a result, the display “Part reject M” lights up.

3. The pulse \(e_x\) is located in time after the emf pulses \(e_m\) and \(e_T\). This case corresponds to a rejected part due to excessive hardness. From Fig. 4, b it is seen that the circuit remains in its initial state, since the coincidence circuits are closed at the first input at the moments when the pulses \(e_m\) and \(e_T\) arrive.

To return the triggers to the initial state, a “Reset” button is provided.

The instrument circuit is assembled on P15 transistors.

LITERATURE

1. I. G. Leshchenko, V. A. Malin. Installation for sorting parts by coercive force. GOSINTI, Instruments and Devices for Defectoscopy, No. 2-64-282/8, 1964.