Guide
Rockwell vs Brinell vs Vickers: which hardness scale should your drawing call out?
The hardness scale is the test method a drawing names next to a hardness number, and the choice of Rockwell vs Brinell vs Vickers decides whether the finished part can actually be tested to the print. Each scale needs a minimum thickness, a flat enough spot and a clean surface; pick one the part cannot give it, and the requirement can only be met by testing something other than the part.
By Saola Manufacturing engineering team · Published · Updated
Most hardness callouts are copied from an older drawing or a heat treater’s brochure, and the typical result is a line such as “HARDNESS 40 HRC MIN.” with nothing else. That reads clearly, but it leaves open where to test, before or after plating, how many parts, and what to do when a thin flange or a small pin is too small for a Rockwell C indent. This guide is about choosing the scale so that the number on the drawing is one the part can physically be tested to: steel, aluminum, die castings, thin sections and plated or case-hardened parts. For what a hardness value on a mill certificate does and does not prove, see how to read a mill test report.
On this page
Rockwell vs Brinell: what does each test actually measure?
Rockwell measures how deep an indenter sinks between a light preload and a heavy load, so it reads directly off the machine. Brinell and Vickers measure the size of the dent left by a ball or a diamond pyramid under a known force, so the indentation has to be measured optically.
That difference drives almost every practical choice. Rockwell is fast, needs little surface preparation and suits production checks on hardened steel, which is why it dominates shop drawings. Per ASTM E18, a regular Rockwell number is 100 minus the permanent depth divided by 0.002 mm for diamond scales such as HRC, and the superficial scales (HR15N, HR30N, HR15T, HR30T) use 0.001 mm per point under lighter loads. Brinell, governed by ASTM E10, presses a tungsten carbide ball, most often 10 mm, into the surface and averages a large impression, which is exactly what you want on coarse or uneven structures such as castings and forgings, and exactly what you do not want on a small finished part. Vickers, per ASTM E92 and the microindentation method ASTM E384, uses one diamond pyramid across a wide range of forces, so the same scale covers a heavy forging test and a traverse through a hardened case.
| Scale | Indenter and total force | Typical use | Minimum thickness rule |
|---|---|---|---|
| HRC | 120° diamond cone, 150 kgf (1471 N) | Hardened and tempered steel, roughly 20 to 70 HRC | At least 10× indent depth; about 0.8 mm at 60 HRC, 1.4 mm at 30 HRC |
| HRBW | 1/16 in (1.588 mm) carbide ball, 100 kgf (981 N) | Annealed steel, brass, harder aluminum | At least 15× indent depth; about 2.1 mm at 60 HRBW |
| HR15N / HR30N | Diamond cone, 15 or 30 kgf | Thin hardened parts, shallow cases | About 0.2 mm at 80 HR15N |
| HR15T / HR30T | 1/16 in carbide ball, 15 or 30 kgf | Thin soft steel, brass and aluminum strip | Several tenths of a millimeter, by hardness |
| HBW 10/3000 | 10 mm carbide ball, 3000 kgf (29.42 kN) | Steel bar, forgings, cast iron, up to 650 HBW | At least 10× indent depth; about 4 mm at 250 HBW |
| HBW 10/500 | 10 mm carbide ball, 500 kgf (4.90 kN) | Aluminum and zinc die castings, copper alloys | At least 10× indent depth |
| HV1 to HV30 | 136° diamond pyramid, 1 to 30 kgf | Any metal, small parts, thin sections | At least 1.5× the indent diagonal |
| HV0.1 to HV0.5, HK | Diamond pyramid or Knoop, 100 to 500 gf | Case-depth traverses, coatings on a cross-section | Tested on a mounted, polished section |
Which hardness scale fits hardened steel, soft steel and cast iron?
Use HRC for through-hardened or quenched and tempered steel, HRBW for annealed or free-machining steel, and Brinell for bar, forgings and cast iron. Switch scales instead of quoting a number outside the range the scale is valid for.
Rockwell C is the default for heat-treated steel, but the scale has a floor and a ceiling. Per ASTM E18 the HRC scale is applicable from about 20 HRC to about 70 HRC; a callout such as “15–20 HRC” asks for readings in a zone where the diamond scale loses resolution, and the part should be specified in HRBW or Brinell instead. The opposite mistake appears on soft steel: the B scale tops out at 100 HRBW in ASTM E18 and ISO 6508-1, so readings that climb toward that limit mean the material belongs on the C scale.
Brinell is the honest choice when the material is coarse-grained or the hardness matters through a section rather than at a point. Steel bar specifications and forging purchase orders often state hardness in HBW for that reason, and gray or ductile cast iron is almost always specified in Brinell, because a small Rockwell indent can land on a graphite flake and read low. The limit, per ASTM E10, is 650 HBW with a tungsten carbide ball; above that, Brinell no longer applies and the part needs Rockwell C or Vickers. Brinell also needs room: a 3000 kgf impression on a typical quenched and tempered shaft is several millimeters across, too large for a small turned part and too visible for a cosmetic face.
Free-machining grades deserve a note. Leaded and resulfurized steels such as 12L14 are bought for machinability, not strength, and are usually supplied cold drawn; if the drawing carries a hardness number at all, it is typically an HRBW range used to confirm the bar condition. The grade choice itself is covered in 12L14 vs 1215 free-machining steel.
Which hardness scale fits aluminum and die castings?
For wrought aluminum, treat hardness as a screen for temper, not as an acceptance property, and use HRBW, HRE, HR15T or a Webster gauge depending on thickness. For die castings, Brinell with a 500 kgf load per ASTM E10 is the usual choice because it averages over porosity and skin.
Wrought aluminum product standards define tempers by tensile and yield strength, not by hardness, so a hardness number on a 6061-T6 drawing cannot by itself prove the temper. It is still useful as a quick check that a part was not left annealed or overheated in a later process. Typical handbook values put 6061-T6 near 95 HB (500 kgf, 10 mm ball), which sits in the lower part of the HRBW range; on thin material, a superficial ball scale or the Webster gauge of ASTM B647 fits better. When temper really matters, pair hardness with eddy-current conductivity per ASTM E1004. How T6 and T651 differ despite reading alike is covered in What is the difference between 6061-T6 and 6061-T651?.
Die castings are a different problem. The skin of a high-pressure die casting is denser than its core, and the core carries gas and shrinkage porosity, so a small Rockwell or Vickers indent can read high on the skin and low over a pore. A Brinell impression samples a larger area and scatters less; typical published values for A380 sit around 80 HB. Remember that die-casting alloy standards treat hardness as a typical property, not a guaranteed one, so put a hardness requirement on a casting only when the part genuinely needs it, and expect it to be agreed at first article. Heat treating ADC12 to raise hardness has its own risks, described in ADC12 heat treatment.
How thin, small or curved can a part be before a hardness test stops being valid?
Every method needs material well beyond the indent: per ISO 6508-1, Rockwell needs at least 10 times the indent depth with a diamond and 15 times with a ball, and per ASTM E92, Vickers needs at least 1.5 times the indent diagonal. If the anvil leaves a mark on the back of the part, the reading is invalid.
Run the numbers before choosing the scale. Per ASTM E18 an HRC reading of 60 corresponds to a permanent depth of 0.08 mm, so the section under the indent needs to be at least 0.8 mm thick; at 30 HRC the depth is 0.14 mm and the minimum is about 1.4 mm. By the same ISO 6508-1 rule, a typical stamped 0.5 mm hardened spring clip cannot be tested on HRC at all, while the superficial HR15N scale, with roughly an eighth of the depth at that hardness (about 90 HR15N per the ASTM E140 steel table, so about 0.01 mm), can. For soft thin strip, the HR15T and HR30T ball scales or a low-force Vickers test do the same job. A Vickers HV10 indent on a 200 HV part has a diagonal of about 0.30 mm, so per ASTM E92 it needs only about 0.46 mm of material, which is why Vickers is the fallback for anything small.
Edges and spacing matter as much as thickness. Per ASTM E18 and ASTM E10, the center of an indent should sit at least 2.5 indent diameters from an edge and 3 diameters from the next indent; on light metals, ISO 6507-1 asks for wider spacing for Vickers. A narrow web or a small boss may be too close to an edge for any regular Rockwell test.
Curvature is the third trap. A convex cylinder reads softer than a flat on Rockwell, and ASTM E18 publishes correction values by diameter; ISO 6507-1 does the same for Vickers. On small turned pins the correction becomes large enough that a ground flat, an end face or a test coupon from the same heat treat load is a cleaner answer. Whichever you choose, mark it on the drawing so the test location is not left to whoever picks the part up.
How do you specify hardness on a plated, anodized or case-hardened part?
Specify substrate hardness on bare metal, before plating or on a masked spot, and specify a case by depth and a Vickers traverse rather than by a surface Rockwell number. A coating or a thin case changes what a surface indent measures.
A zinc or nickel plating layer is thin compared with a regular Rockwell indent, but it is not invisible: soft zinc can lower a superficial reading, and a hard plating can raise one. The clean callout is “hardness verified after heat treatment, before plating”, with the plating requirement stated separately. If the coating itself must be hard, as with hard anodize or electroless nickel, the only meaningful test is a microhardness reading on a polished cross-section per ASTM E384, and that is a destructive test on a sample, not a check on every part. Hardness also interacts with plating chemistry: high-strength steel picks up hydrogen in acid cleaning and electroplating, which is why the hardness of the part sets the bake requirement described in hydrogen embrittlement baking after zinc plating. For anodize, the coating types are compared in Type II vs Type III anodize.
Case-hardened parts need two numbers: surface hardness and case depth. A surface HRC reading on a shallow carburized or nitrided case pushes through the case into the softer core and reads low, so use a superficial or Vickers scale for the surface. Define the depth as effective case depth, for example to 50 HRC equivalent per SAE J423, measured by a Vickers or Knoop traverse on a sectioned sample from the same load.
Can you convert between Rockwell, Brinell and Vickers numbers?
Only approximately, and only with a table made for that material. Conversions per ASTM E140 are empirical, so a converted value can guide a decision but should not be the acceptance basis.
ASTM E140 publishes separate conversion tables for nonaustenitic steel, nickel alloys, cartridge brass, austenitic stainless, copper, alloyed white iron and wrought aluminum, and it states that each table is valid only for the material it names. Two common failures follow. A typical case: a drawing states “30 HRC (286 HB)”, and two labs measuring the same bar disagree on which number governs; the fix is to make one scale the requirement and mark the other “REF”. Or an aluminum part is specified in HRC because a steel table offered a conversion, which gives a number nobody can test; aluminum has its own E140 table, on its own scales. If a supplier measures on a different scale for practical reasons, agree that in writing at first article and keep the reading on the scale actually used.
How should the hardness callout read on the drawing and the RFQ?
State the range, the scale, the test standard, the test location and the stage, and say who tests and what the report shows. The typical bare “40 HRC MIN” leaves every one of those to the supplier.
Examples that can be tested as written, per the standards named in each:
HARDNESS 28–34 HRC PER ASTM E18, ON END FACE AT FLAG 1, AFTER HEAT TREAT AND BEFORE PLATING.HARDNESS 229–285 HBW 10/3000 PER ASTM E10, ON MACHINED SURFACE, ONE READING PER HEAT TREAT LOAD.SURFACE 653 HV1 MIN PER ASTM E92 (58 HRC REF PER ASTM E140); EFFECTIVE CASE DEPTH TO 513 HV (50 HRC) PER SAE J423, VICKERS TRAVERSE ON SECTIONED SAMPLE.6061-T6: TEMPER PER MATERIAL SPECIFICATION; HARDNESS REF ONLY, PER ASTM B647 OR ASTM E18 HRBW.
Keep ranges realistic. A production quench and temper does not land on a single HRC point, so give a band several points wide; a single-point value or a very narrow window tends to be either scrapped or quietly widened. On the RFQ, list the material and its condition as delivered, the heat treatment if any, the hardness range and scale, the test location, the sample size per lot, and whether you want actual readings on the report or a pass statement. Name the tester too: the heat treater, the machining supplier, a third-party lab or your own receiving inspection. Hardness belongs on the first article inspection report and the certificate of conformance like any other drawing requirement.
To be plain about the plant behind this site: the quality page lists first-article inspection reports, dimensional inspection records with each lot and material certification on request. It does not list a hardness tester or a hardness test report, and heat treatment is not among the processes the factory page lists as run in-house. The bar materials listed for CNC turning include 52100 and 12L14 steel and 2011-T3 and 6061-T6 aluminum; the surface finishing menu includes zinc plating, electroplating and anodizing on aluminum, and that page asks for any finish or test standard it does not run to be named before tooling. If your part carries a hardness requirement, put the scale, location, stage and tester on the RFQ so the requirement is quoted rather than discovered.
Frequently asked
Is Rockwell or Brinell more accurate?
Neither is more accurate in general. Brinell averages over a large area and suits coarse or uneven material; Rockwell is faster and suits small, uniform, hardened parts. The right one is the one the part can carry without breaking the thickness, edge and curvature rules.
Does a hardness test damage the part?
It leaves a permanent indent. On a sealing face, bearing seat or cosmetic surface that indent can be a defect, so name a test location on a non-functional face or call for testing on a coupon or a sacrificial part.
Why do two labs report different hardness on the same part?
Usually because they tested different spots, used different scales and converted, tested through plating or decarburized skin, or ignored a curvature correction. Fixing the scale, the location and the surface condition on the drawing removes most of the disagreement.
Can a mill certificate replace a hardness test on finished parts?
Only when nothing after the mill changes the hardness. Any heat treatment, welding, cold work or plating bake after the bar leaves the mill means the certificate describes the raw material, not your part.
What scale suits a zinc die casting?
Brinell with a low load is the usual choice, for the same reason as aluminum castings: it averages over skin and porosity. Zinc alloys also age at room temperature, so state when after casting the reading is taken.
Sources
- ASTM E18, Standard Test Methods for Rockwell Hardness of Metallic Materials, ASTM International, current edition, accessed October 2026
- ASTM E10, Standard Test Method for Brinell Hardness of Metallic Materials, ASTM International, current edition, accessed October 2026
- ASTM E92, Standard Test Methods for Vickers Hardness and Knoop Hardness of Metallic Materials, ASTM International, current edition, accessed October 2026
- ASTM E384, Standard Test Method for Microindentation Hardness of Materials, ASTM International, current edition, accessed October 2026
- ASTM E140, Standard Hardness Conversion Tables for Metals, ASTM International, current edition, accessed October 2026
- ASTM B647, Indentation Hardness of Aluminum Alloys by Means of a Webster Hardness Gage, ASTM International, current edition, accessed October 2026
- ASTM E1004, Determining Electrical Conductivity Using the Electromagnetic (Eddy Current) Method, ASTM International, current edition, accessed October 2026