Abstract
          	The aim of the submitted work is to evaluate hardness measurement process capability of two test methods Brinell (HBS5/250/30) and Vickers (HV10/30).  
           	The Metals Handbook defines hardness as "Resistance of metal to plastic deformation, usually by indentation. However, the term may also refer to resistance to scratching, abrasion, or cutting. It is a property of a metal, which gives it the ability to resist being permanently deformed (bent, broken, or have its shape changed), when a load is applied. The greater the hardness of the metal, the greater resistance it has to deformation. Measurement of the macro-hardness of materials is a quick and simple method of obtaining mechanical property data for the bulk material from a small sample. 
           	The Brinell hardness test uses a machine to press a 2.5, 5 or 10 mm diameter, hardened steel (or tungsten carbide) ball into the surface of the test specimen. The machine applies a load of 500 kilograms for soft metals such as copper or brass. A 1500 kilogram load is used for aluminum castings, and a 3000 kilogram load is used for materials such as iron and steel, all for ball diameter 10 mm. The load is usually applied for 10 to 15 seconds, alternatively 10-180 seconds for soft metalls. Among used hardness tests, the Brinell ball makes the deepest and widest indentation, so the test averages the hardness over a wider amount of material, which will more accurately account for multiple grain structures, and any irregularities in the uniformity of the alloy (typical for cast structure) and soft materials. A wide indentations, on the other hand, can impair the surface of specimen. 
          	The Vickers test is the standard method for measuring the hardness of metals, particularly those with extremely hard surfaces: the surface is subjected to a standard pressure for a standard length of time by means of a pyramid-shaped diamond. The diagonal of the resulting indention is measured under a microscope. The Vickers testing method is similar to the Brinell test, it is the most accurat and sensitive hardness test method. It is unsuitable for inhomogeneous and coarse-grained materials. Thoroughly prepared surface befor test is unavoidable. 
          	Attempts to standardise evaluation of the measurement processes capability can be observed for several years.  The capability of measurement processes is searched  similiarly to that  of production processes. Process capability means the ability of the process to meet technological or other requirements.  Measurement process capability is determined by total variation caused by random reasons influencing the process. 
          	Variability of the measured data (caused by the measurement process) as well as the systematic deviation from the required values is observed during the evaluation of the measurement process capability.  Both such efforts are determined by the technological development of the measuring instruments. Due to the technical innovations (incorporating electronics, autocalibration, automatic correction of the output signal according to the status of the measuring instrument) the variability of the measurement process is decreasing and the effort for its centering improves the capability of measuring instruments. 
          	Measurement system analysis (MSA) is an experimental and mathematical method of determining how much the variation within the measurement process contributes to overall process capability. MSA involves gauge repeatability and reproducibility (R&R or GRR) studies to evaluate measurement systems. It is an interactive, multimedia, system designed to help engineers and quality professionals assess, monitor, and reduce measurement system variation. MSA helps us conform to ISO 9000 and ISO/TS 16 949:2002 requirements. 
          	To conduct a R&R study, you will need the same type of parts (25 “fieds” in our experiment), at least two appraisers (people who measure the parts – 3 in our experiment), one measurement instrument or gauge and a minimum of two measurement trials (5 in our experiment), on each part, by each appraiser.  We need multiple trials to estimate repeatability %EV and multiple operators to estimate reproducibility %AV. Multiple parts allow us to obtain better estimates of repeatability and reproducibility, as well as to estimate part variation %PV.  Index  %R&R (or %GRR) represents the overall process capability.
          	Al-Si cast slab with 9.75 wt% Si, 0.4 wt% Mg, 0.4 wt% Mn and 0.16 wt% Fe (STN 42 4331) was the experimental material.  Final Sr content 0.015 wt% was sufficient for real modification of eutectic Si. The pouring temperature was 760°C. The melt was poured into the chill mould, the casting cooling rate was 2 °C.s-1 between 760 and 350°C. 
          	The microstructure was evaluated at the longitudinal axis of the castings. Polished metalographic samples were etched with 25% H2SO4 at 75°C and consequently with 0.5% HF. The morphology  of  eutectic Si (?? phase) and its interparticle spacing???,  dimensions and shape of intermetallic iron containing phases as well were evaluated according to the STN 42 0491.
          	The upper surface of slab was milled, ground by abrasive papers No. 220, 240, 280, 400 a 500 and divided into 25 “fields”. The hardness HBS 5/250/30 was evaluated according to ISO 6506 (STN 42 0371). The loading and ball diameter were chosen according   STN 42 4341. The hardness HV10/30 was evaluated according to the ISO 6507-1 (STN 42 0374). 
           	The length of aluminium matrix dendrites was up to 300 ?m in the field No. 1 (the rim of the slab) toward its centre it increased to 500 ?m  in the field No. 4 and 900??m in the field No. 13. The width of dendrites is 22-34 ?m in the field No. 1 and  51-76 ?m in the field No. 13.  The diameter of eutectic silicium  is 1-7 ?m and its interparticle spacing   ?? 2.46 ?m, in field No 1 and up to 10 ?m with interparticle spacing   ?? 3.8 in the fields No 13-16, where the needles of eutectic silicium (up to 20 ?m) can be observed. Rare grey – rusty needles of intermetallic phase  Fe2SiAl8 (phase ?) with the length between 5 and 30 ?m  can be seen in eutectic cells.
          	As a rule of thumb, the measurement system should have resolution  at least by 1/10th smaller than the process variability. This demand is fulfiled, because the resolution is 0.33 HV10/30 and the standard deviation is 4.16HV10 or 0.15 HBS 5/250/30 and  equivalent standard deviation is  4.3 HBS 2/250/30.. 
          	The variation derived from the measurement instrument, method and (unstable) measurement conditions  %EV = 59.5% for HB  and 78.5% for HV.  The variation derived  from the competence of appraisers %AV = 38.1% for HB and 15.3% for HV. The variation derived from the parts %PV = 70.8% for HB and 60.1% for HV.  As   %R&R exceeds 30% (70.6% for HB and 79.9% for HV), the used hardness measurement system as well as process. The inhomogeneity of tested material is causal relation of low capability.