Increase in reliability and resource of machinery is very important condition of creation of competitive products. One of the important conditions of providing quality product is control of mechanical properties of materials and, in first turn, of rolled metal- most widely used structural material. Existing international and local standards stipulate mechanical properties determination as preparation of specimens and destruction test on tensile-testing machine carried out on these specimens. However, selective check of mechanical properties, using statistical criteria, becomes effective only in quantity or mass serial production for which these criteria were established. For unstable production or for production of small batches selective checks are not reliable, and for piecework destructive check is meaningless. For different kinds of transport metal, like wheels, rings, rail, pipe for gas lines, ropes it looks quite necessary to develop automatic system which implements fast non-destructive way for determination of the mechanical properties.
Different methods for sampleless mechanical properties determination are well known: ultrasonic, electro-magnetic, using indenters. The last ones are the most promising, because the character of loading and stressed state is very close to that standards specify. Results of measurements do not depend on chemical composition of the material. Methods of testing which use indenter to load the material, both statically and under shock loading, were developed by I.Brinell, M.Drozd, M.Markovets, E.Meyer, Y.Fridman.
Method described in this paper includes shock impact loading of the object, optico-electonic registration of indenter motion path, digital data processing and determination of mechanical properties of the object. Optico-electronic sensor was selected to track indenter's trajectory to eliminate any mechanical resistance and to obtain very high precision of data recording. Comparable analysis of different types of displacement sensors in an operating range of 1 mm and velocities of 3-7 m/s showed the most suitable sensor for this particular application is light diode - phototransistor optical coupled pair working in non-saturated mode. Optical channel is interrupted by the shutter located on the indenter. Indenter displacement is determined using sensor output signal. Velocity of the indenter is determined based on the time while optical channel is interrupted and shutter dimensions. The contact signal is used to determine the moment when indenter touches material and rebounds out of the surface.
The method for automatic dynamic calibration of the displacement sensor has been worked out in order to linearize and stabilize sensor's characteristics. Structural redundancy was used - second shutter was installed after the first one. Considering velocity constant while shutters move along the optical channel, static characteristic "sensor displacement-sensor output" can be established, and then linearized and stabilized. Signal generated by the second shutter is used to record indenter path in the material. Velocity is determined using time between two shutter's leading edges generated while indenter flies down and up. Described sequence is performed separately for each stage of indenter motion: on penetration and rebound. This allows to reasonably compensate influence of the sensor's delay time on measurement result. The same computer used for main calculations performs calibration.
The main advantage of using automatic dynamic calibration is that it is performed every time measurement starts. There is no need in additional calibration measure or bench; the device is always ready to work. Tests showed that after automatic dynamic calibration non-linearity of the static characteristic doesn't exceed 0,4% (initially up to 15%), instability of linearized charteristic is less than 0,5% with 10% luminous flux change.
Method was tested in a laboratory conditions using special device, which contains shock impact unit, sensor described above, ADC and control computer. ADC resolution allows to determine indenter position with resolution up to 1 mkm.
Impact energy, indenter mass and sphere diameter are chosen taking into consideration metal surface roughness. Metal should be exposed to resilience and plastic deformation during the test. Data processing algorithm uses experimental database, collected during standard and shock impact test carried out on the same material (Proceedings of 7th EC NDT, 1998, v.1, pp.1070-1077). Algorithm is illustrated on fig. 1. It shows creation and analysis of the database.
Method accuracy was examined on the specimens made from X70G steel and construction steel grades. Specimens were subjected to shock impact test and standard strength test. Deviation in YS and UTS values was between 10-20 MPa correspondingly, A4 - 1,5%.
To check hardness number on steel following methodology was developed. Usually to determine maximum contact force piezo accelerometers are used. In general their usage should be avoided because of very low precision (about 10 to 20%) of measurements of shock acceleration. Also very complex calibration benches are required to calibrate and regularly check that sensors. Based on a research of the penetration process, it was proposed to determine maximum contact force using second phase of the process - rebound. This phase could be very precisely described using Herz's formula.
P = K(a -h)3/2.
Vertical coordinate is calculated from the depth h - deformation of the material. Three options are possible to calculate K coefficient: if E and m of the material are known; the same but taking into consideration depth of residual deformation is known; if E and m are unknown, but starting velocity (n = 0) , velocity at the rebound moment (n = nrebound ), indenter weight and elastic recovery w = amax - h are known.
Maxim contact force could be measured without using of accelerometer, using just natural dependency P(a,h). This allows reasonably increase precision and simplifies design of the measuring unit. In order to determine dynamic hardness information about only one parameter is required - indenter coordinate. In order to convert dynamic hardness into static velocity is also required. The same parameters and rebound velocity are required to use automatic identification of mechanical properties.
After some improvements method can be used for express tests during production of railway wheels, tyres, thick plates, rails, beams etc. It also can be used in handheld devices for high-stressed steel constructions tests, like high-pressure and high-temperature pipelines, where mechanical properties may degrade with time.
