Hardness is the number a drawing falls back on when it has to say, in one figure, whether a part was heat treated correctly. The Vickers test is the method that produces that figure across the widest range of materials and the smallest features, from a case-hardened gear tooth to a thin coating to a weld and the heat affected zone beside it. It is also the method where the number depends least on the machine and most on how carefully somebody measured two diagonals.
This is how the test actually works: the geometry behind the HV number, the three ranges of force it runs at, the standards that govern it, and the places where the accuracy is genuinely won or lost.
The One Indenter and the Number It Produces
Every Vickers test on every material uses the same tool: a diamond in the shape of a right pyramid with a square base, with an angle of 136 degrees between opposite faces. That single geometry is the reason the method behaves the way it does. Because the indentation is always the same shape, indentations made at different forces stay geometrically similar, so one continuous scale covers soft aluminium and hardened tool steel without a change of indenter or a correction factor.
The hardness value is the test force divided by the sloping surface area of the indentation the diamond leaves. The operator measures the two diagonals of the square impression, averages them, and the machine turns that average into a number.
Scroll sideways to read the full diagram
HV is a force divided by an area, but it is not a pressure and it is not interchangeable with a stress in pascals. It is written with the force attached, so HV10 is a value taken at 10 kgf and HV0.5 one taken at 500 gf, and a bare HV number without its load is incomplete.
That last point is worth holding on to, because it explains the whole error story further down. Hardness goes as the inverse square of the diagonal. A small mistake in reading the diagonal becomes a larger mistake in the reported hardness, and the smaller the indentation, the more that leverage hurts.
How a Vickers Test Runs
The sequence is short, and each step exists because skipping it changes the number.
The cycle
Four steps, and the last one decides the number
Three of these steps are preparation. The measurement itself happens only at the end, after the force has already been removed.
Prepare the surface
The diagonals have to be visible under a microscope, so the test area is ground and polished to a metallographic finish. Smearing left by grinding sits exactly where the diamond will land.
Apply the force under control
The force is brought up smoothly, not dropped. A modern tester applies it through a load cell or a dead weight system and holds it steady.
Hold for the dwell
Materials that creep keep flowing while the force is on, so a longer hold on such a material yields a larger indentation and a lower hardness. The dwell is part of the result.
Measure both diagonals
Only after the force is removed. The operator, or a camera and software, reads the recovered impression. This is the measurement; everything before it was preparation for it.
The force is never dropped and gravity is not what drives the diamond in. That distinction matters, because a machine that applied force by impact would be measuring something else entirely, and there are portable methods that do exactly that.
Surface preparation is the step most often underestimated. In practice it means the same progression used for metallographic samples, which is covered by ASTM E3 and described in our guide to metallographic sample preparation. On a production part that is a real operation with a real cycle time, not a wipe with a cloth.
A full Vickers and Knoop cycle on an NG-1000, from placing the sample to reading the diagonals.
Three Ranges of Test Force, One Scale
Vickers is often described as two methods, micro and macro, but it is one method run at different forces. ISO 6507-1 divides it into three ranges rather than two, and the boundaries are set in newtons rather than in the kilogram force units the scale is usually quoted in.
The force decides how much material the indentation averages over, and that is a metallurgical choice before it is an equipment choice. At the bottom of the range the indentation lands inside a single grain or a single phase and reports that feature. At the top it spans a coarse structure and reports the bulk. Neither is more correct; they answer different questions, and specifying the wrong one is how a casting gets rejected on the hardness of one carbide.
Scroll sideways to read the full diagram
Boundaries and hardness symbols are taken from ISO 6507-1, Table 1. The ASTM figures come from ASTM E92 and ASTM E384. Because the two systems place their boundaries differently, a report that says only “micro Vickers” without naming the standard has not said very much.
Because the same scale runs across all three ranges, a value only means something with its force attached. HV0.1 and HV30 taken on the same sample can legitimately differ, and on a fine or layered structure they usually do. Our overview of macrohardness against microhardness testing goes into where that split lands in practice.
Where the Accuracy Is Won or Lost
The Vickers machine applies a force and holds it. That part is straightforward, well controlled on any competent instrument, and rarely the problem. The measurement of the diagonals is where the number is decided, and the published metrology work is unusually blunt about it.
The UK National Physical Laboratory surveyed fifteen commercial hardness machines against its own national standard machines, on the HV10 and HV30 scales at nominal block hardnesses of 150, 300 and 800 HV. Repeatability was excellent. Even so, over half the values fell outside the two percent tolerance, and correcting the results for systematic errors and for the magnification of the microscope used to measure the diagonal brought the uncertainty back inside it. No single mechanical cause was found. Over half of the commercial measuring systems in that survey exceeded the permitted half percent magnification tolerance.
Read that chart next to the inverse square relationship and the conclusion is uncomfortable but useful. On a large macro indentation a two micrometre disagreement between two people is a rounding error. On a small micro indentation it is a double digit percentage of the hardness value, and no amount of machine calibration recovers it, because the machine did its job correctly and the disagreement happened afterwards, at the eyepiece.
This is the single strongest argument for taking the reading off a camera and software rather than a human eye, and for keeping certified reference blocks in the routine. It is also why the optics are the expensive part of a Vickers tester rather than an accessory to it.
The Limits the Standard Sets on the Method
ISO 6507-1 does not simply describe the test, it fences it. The test method is specified for indentation diagonals between 0.020 mm and 1.400 mm, and determining Vickers hardness from indentations smaller than that lower limit is stated to be outside the scope of the document. That is not a formality. It is the same physics as the chart above, written into the standard: below a certain size the diagonal cannot be measured well enough for the result to mean anything.
The practical consequence is that the lower end of the force range is bounded from two directions at once. The standard permits forces down to 0.009807 N, but it also requires that the force chosen produces a diagonal longer than 0.020 mm. On a hard material a very small force produces a very small indentation, and the two requirements collide. The force is therefore selected from the material and the feature together, not from the feature alone.
Conditions that travel with the number
Three things a Vickers result carries whether or not they are written down
Temperature
Tests carried out under controlled conditions are made at 23 ± 5 °C. Hardness is a mechanical property of a material at a temperature, and a laboratory that reports to tight tolerances without controlling the room is reporting a wider result than it thinks.
Spacing between indentations
An indentation work hardens the material around it, so a second one placed too close reads high, and one placed too near an edge reads low because the material there can move. The standards set minimum centre to centre spacings and edge distances in multiples of the diagonal, which is why a hardness traverse across a case depth is laid out on a measured pitch rather than by eye.
The force itself
A Vickers value is designated with its force, and the designation is part of the result. HV 30 and HV 0.5 are not two ways of writing the same measurement. On a homogeneous material they will agree closely; on a layered or fine structure they are answering different questions.
Choosing the Instrument
Once the force range is decided by the metallurgy, the instrument follows from two questions: how small the feature is, and how many indentations a shift has to produce. A single spot check and a two hundred point case depth map are the same test method and very different machines.
Vickers and Knoop
Three instruments, three force ranges
All three read the diagonals optically. What separates them is the force they can apply and how much of the reading they take off an image rather than an eyepiece.
Micro
NG-1000 Micro Vickers and Knoop
Loads from 10 g to 1 kg, motorized turret as standard, in five configurations from an eyepiece read digital tester up to a fully automatic system with a CCD camera and a motorized XY stage.
Case depth profiles, thin coatingsIndividual phases in a microstructure
Loads from 300 g to 50 kg, motorized turret, large display and a built-in printer, with an optional CCD camera and analysis software that reads the diagonals from an image instead of the eyepiece.
Castings, welds and heat affected zonesHeat treated bulk parts
Vickers and Knoop, Rockwell and Brinell on one frame, with an 8 inch touchscreen, a CCD camera and automatic measuring software, supplied with NIST certified consumables for all three methods.
Several methods, one benchOne instrument to calibrate rather than three
Automation changes the economics more than it changes the physics. A motorized turret removes the manual swap between indenter and objective after every indentation. A motorized stage plus analysis software lets an operator lay out a pattern of indentations, start it, and come back to a plotted result, which is what turns a case depth survey from an afternoon into a background task. The whole Vickers and Knoop range shares the same measurement principle underneath.
The fully automatic configuration running a programmed pattern of indentations without an operator at the eyepiece.
What Has to Be Right Besides the Machine
A Vickers result is only defensible if three things around the machine are in order, and all three are consumable or perishable in some sense.
The diamond comes first. It is the one part of the system in contact with every test, and its geometry sits inside the hardness equation rather than beside it: across the angle band the standard permits, measured hardness on the same block shifted by around four HV. A chipped or worn indenter does not announce itself in the reading, which is why indenter condition is checked rather than assumed.
Certified reference blocks come next. They are how a laboratory demonstrates on any given morning that the whole chain, force, diamond, optics and operator, still produces the right answer. NIST supplies microindentation hardness reference materials for exactly this purpose on the Vickers and Knoop scales, and NIST certified indenters and test blocks are available with our systems on request.
The optics and the software come third, and by the evidence above they carry most of the risk. A camera and analysis software do not make the method more accurate in principle. They remove the operator to operator spread that the metrology studies keep measuring, and they make the reading reproducible by somebody else six months later, which is usually what an audit is asking about.
Vickers Against Brinell, Rockwell and Knoop
Vickers earns its place through range and locality rather than speed. One indenter and one continuous scale cover materials that would need several Rockwell scales, and the indentation can be placed precisely enough to interrogate a single feature. Against that, the cycle is slower than Rockwell, which reads depth directly and gives a number in seconds without any optical step, and the surface preparation is far more demanding.
Against Brinell, Vickers works on much harder material and much smaller features, while Brinell’s large ball averages over coarse and inhomogeneous structures like castings in a way a small diamond cannot. Knoop, which shares the same machines and the same optical workflow, uses an elongated rhombic indenter whose long diagonal makes it better suited to very thin layers and brittle materials. If the choice is still open, our comparison of Brinell, Rockwell, Vickers and microhardness works through the decision by application.
Reading a Vickers Result on a Drawing
A hardness callout that says only a number and the letters HV is incomplete, and the ambiguity it leaves is not academic. The designation carries the force, and where the dwell differs from the normal range it carries that too, so a specification written as HV 10 and one written as HV 0.3 are asking the laboratory for two different tests on two different scales of feature.
Two habits cause most of the trouble in practice. The first is treating an HV number as convertible into another scale as though the conversion were exact. Conversions between hardness scales are empirical, they depend on the material, and a converted value is an estimate rather than a measurement; where a specification is written in one scale, the honest answer is to test in that scale. The second is quoting a hardness without the standard it was produced under, which, given that ISO and ASTM place their range boundaries in different places, can leave two laboratories running genuinely different tests against the same line on a drawing.
The fix is unglamorous and takes one line. State the scale with its force, name the standard and its part, and say where on the part the measurement is to be taken. A drawing that does that gets comparable numbers from any competent laboratory, and one that does not will eventually get two numbers that disagree for reasons nobody can reconstruct.
Next step
Not sure which force range your parts need?
Send us the material, the feature you need to measure and the standard you are working to, and we will tell you whether it is a micro, macro or universal instrument, and what consumables it needs to produce traceable results from day one.
Flexural (Bend) Testing Explained: 3-Point vs 4-Point, Standards, and SetupFlexural (bend) testing explained: what it measures, three-point vs four-point setups, the standards for plastics, metals, composites and concrete, and how to choose the right fixtures and...Read the article
Vickers and Knoop Hardness TestersHardness Test Blocks, Indenters and AccessoriesHardness test blocks, indenters, and accessories support daily verification and reliable operation of metal hardness testing systems. The range includes certified...View product