Method for detecting ingot cracks

CLAIM FOR THE PRIORITY of the patent application / dd / Wd'déJ'e'd'Ziiility

In Norway

From November 3, 1981

Brief Description filed in support of a request for

PATENT OF INVENTION

Luxembourg on behalf of: elkem a / s for * Procedures for detecting defects in billets.

DESCRIPTIVE MEMORY

SUBMITTED IN SUPPORT OF A REQUEST

OF

PATENT

TRAINED BY

ELKEM A / S for Methods for detecting defects in billets.

The present invention relates to a detection method and in particular to a method for detecting surface cracks or defects in billets, so that they can be indicated and eliminated, for example - by grinding the billet, so that the defects do not follow the billet in the remainder of the transformation process into finished products.

GM.MJ 4B 7L

E-800

By the term billet, is meant in the present specification both metal parts to be rolled for the first time as parts to be subjected to a new rolling treatment and finished products and in particular bars, tubes, pipes, profiles and other objects. which must undergo a rolling treatment.

It is known that cracks and surface defects in billets can be detected by heating the surface of the billet by means of a high frequency current while the billet is driven lengthwise through a spool of. high frequency induction, the part of the billet which has just left the coil being, while the billet is passing through the coil, immediately scanned by means of infrared ray recording equipment to record the temperature distribution on the surface of the billet. The temperature distribution is recorded and is visualized as a motley pattern that indicates cracks and surface defects in the billet. This recording of the temperature curve on the billet by means of a sweep shows an increase in temperature in the curve near the cracks. The temperature increases which are repeated from one scan to the next then, when combined, form a temperature ridge along the billet which indicates the presence of a crack in this billet. The invention aims to record very shallow surface defects (less than 1 mm in depth) and to obtain uniform emission conditions for all surfaces, while reducing disturbances in the temperature distribution.

With regard to the detection method described, it has been found that temperature measurements carried out by means of an analysis device or bala

- 2 GM. MJ yage of infrared rays (infrared emission pyrometer) give temperatures which deviate from the actual temperature of the surface due to the emission coefficient of that surface. The emission coefficient depends on the nature of the surface and varies between 0 and 1. Rolled billets with rolling scale usually have an emission coefficient of about 0.90 to 0.95.

To obtain the best conditions during detection and to minimize disturbances in the temperature distribution as much as possible, it is desirable to have as clean a surface as possible. Such a surface can be obtained, for example, by subjecting it to shot blasting and the degree of cleaning can, for example, be classified according to Swedish standard SIS 055990/1967. However, an undesirable side effect of this shot peening is to make the surface shiny and to change its emission conditions. This again affects the temperature which is recorded by the infrared emission pyrometer, so that temperature measurements made with such a pyrometer give too low surface temperatures, with the result that too low a temperature in the thermometer. the cracks are recorded. Therefore, in the case of shot-blasted billets, the temperature rise recorded in the cracks is attenuated compared to that of billets with rolling scale, so that the interpretation of the crack depth as a function of of the emission coefficient of the surface is again different. This is an undesirable effect which complicates the use of the method because, in order to obtain correct results, the equipment must be recalibrated for each new type of surface. This recalibration is time consuming, complicated and not suitable for

GM. MJ an industrial operation.

Applicants have now found, however, that the emission factors of all types of billet surfaces can be made approximately equal by wetting the surfaces with a suitable liquid, for example water containing additives to. to reduce the surface tension and to improve the wetting obtained. When subjecting billets to detection at temperatures below 0 ° C, additives can also be added to lower the freezing temperature. Experiments have shown that such wetting of the surface of the billet by means of water to which additives have been incorporated aimed at reducing the surface tension makes it possible to ensure that all surfaces of the billets, regardless of: , have approximately the same emission coefficient, and that this emission coefficient deviates only to a small extent from the emission coefficient of a dry billet comprising rolling scale. The variation in the emission coefficient for wet billets is so small that it is of no practical importance for the result of the detection method described above. The wetting of the billets also helps to attenuate the disturbances in the temperature distribution because the differences in emission coefficients for different areas of the surface are equalized and the signal / disturbance ratio in the signals from the infrared emission pyrometer is considerably. improved. In the case of billets which have undergone shot peening, but which have not been wetted, the signal to disturbance ratio is so low that certain types of billet surfaces cause difficulty in detecting the shallower cracks.

The result of effecGM.MJ killed temperature measurements with a contact pyrometer and an infrared emission pyrometer for different types of billet surfaces, both wet and dry, are shown in the table in Fig. 1 attached.

Test piece A relates to a billet comprising rolling scale. Specimen B is for a billet with a corroded (rusty) surface, while specimens marked CSA 1 2, CSA and CSA 2¾ are shot-blasted with the surfaces defined in accordance with Swedish standard SIS 055900-1967.

In the table, T indicates the temperature read by the contact pyrometer, while IR indicates the temperatures read by the infrared emission pyrometer. As mentioned above, readings are taken on both a dry and wet surface.

As indicated in the dry columns, a significant temperature difference is present between the two measurements and this difference increases with the degree of shot peening which is designated here respectively by CSa 1 2 CSa2 and CSa 2¾. The difference between the two measuring principles therefore increases with the gloss of the billet. As a result, measurements made by means of the infrared emission pyrometer on shot-peened bilettes give the recording too low a billet temperature and therefore a less increase in scale temperature. However, as mentioned above, in order to obtain acceptable results, it may be necessary to calibrate the equipment according to the emission coefficient of the surface, which is a significant drawback because the emission coefficient can vary d 'one billet to another.

The reduction of temperature peaks at the

GM. MJ of cracks for shot peened surfaces also results in working with smaller tolerances between the temperature peaks at the cracks and the flawless surface temperatures, so that a ratio is obtained. reduced signal / disturbance which is unfavorable.

Tests have shown, as mentioned above, that the emission coefficient for different billet surfaces can be changed and can be made approximately equal for all surfaces by wetting these surfaces. By the term approximately equal it is meant that the emission coefficients of the surfaces of the billets fall within an area which can be accepted for the results of the detection process. The billet may be wetted before it passes through the induction coil and the wetting liquid is delivered as evenly as possible over the surface by a nozzle, slit or the like or by gravity flow along the surface. The effect of this wetting is indicated in the table in the columns marked wet.

The measurement results for a dry surface and for a wet surface which are shown in the table are also shown graphically in Fig. 2 attached. These diagrams show that the emission factors for wet surfaces are all at the same level as for a dry billet with rolling scale, billet A (A A7B). It should also be noted that in this case, the wetting makes it possible to obtain surfaces which all have approximately the same emission coefficient which is relatively close to the theoretical maximum emission coefficient equal to 1.

As mentioned above, the wetting of

GM. MJ billet also offers the advantage that the level of disturbance for a defect-free surface defined by temperature variations on that surface is reduced as the temperature rise due to the presence of a defect producing a temperature peak. on the flawless area is increased, which means that the signal / disturbance ratio is improved.