Hardness Limits and Spalling in Striking Tools
A striking tool is the one tool steel application where the hardness that makes the tool useful is also the hardness that makes it dangerous. Past a certain point the striking face stops deforming and starts fracturing, and the piece that leaves the face carries enough energy to injure whoever is holding the handle. This page covers the hardness window the published specifications set for striking and struck tools, what the tool does at each end of that window, how an acceptable tool is proved by test, and what a large spall says about the steel underneath it.
The hardness window is a safety limit, not a wear limit
The maximum hardness permitted for a striking face by two of the specifications the source tabulates is 60 HRC, and the source makes clear why the number is where it is. Brittle cleavage fracture tends to occur at slightly higher hardness than that, so 60 HRC is very close, perhaps too close, to exceeding the safety limit for a striking tool. The boundary between a safe and an unsafe upper hardness is not discrete but transitional, which is why the same figure appears in several specifications as a shared judgement rather than as a sharp threshold.
The lower limit works the other way. A striking face is usually held at about 50 HRC or above, and where the tool can be used to strike another steel tool such as a wedge, the lower limit drops to about 45 HRC. That figure is chosen for the warning it gives. At 45 HRC the strain hardening exponent is greater and the deformation is more widespread, so the tool tends to mushroom instead of breaking, and the mushroomed face tells the user that a spall is coming. A face that is too hard to mushroom is also a face that gives no notice.
The struck tool carries the same logic in the opposite direction. The maximum allowable hardness of a struck tool is lower than that of the striking tool, so the struck tool takes the macroscopic plastic deformation, its contact area grows, and the stress in the striking tool is spread over a greater volume. A soft struck tool is therefore part of the protection of the striking tool, and the mushrooming it develops is deliberate.
| Hardness of the striking face | What the tool does under a blow | What it means for the user |
|---|---|---|
| Above the permitted maximum | Brittle cleavage fracture starts at slightly higher hardness than the specifications allow | The piece that leaves the face carries enough energy to injure, which is why two of the specifications stop at 60 HRC and the source calls that figure very close to the safety limit |
| At the top of the range, 55 to 60 HRC | The face stays flat and does not deform before it breaks | Wear resistance at its best and warning at its worst. A face that cannot mushroom gives no sign before it spalls |
| Mid range, 50 to 55 HRC | The face wears and deforms slowly under repeated blows | The practical band for a hammer head, a percussion tool and a heavy striking tool |
| About 45 to 50 HRC | The strain hardening exponent is greater and the deformation is more widespread, so the face mushrooms | The mushrooming is a warning that spalling may be imminent. This is also the band used where the tool has to strike another steel tool such as a wedge |
| Struck tool, below the striking tool | The struck tool deforms first and its contact area grows | The larger contact area spreads the stress over a greater volume of the striking tool, so a soft struck tool is the safety measure and not a compromise |
The hardness window for a striking face and what the tool does at each point in it. Compiled from the comments on specifications and the striking test discussion in the article Spalling from Impact Events.
What the specifications actually require
The specifications differ in more than their hardness ranges. They differ in the steel they name, in which face of the tool they control, and in the test used to prove the tool. A percussion track tool is controlled at four places, the striking face, the struck face, the cutting edge and the point end, with a different hardness allowed at each. A hammer is controlled at the face and again three millimetres below the face, which is where a hard surface over a soft interior would otherwise hide. A chisel is allowed a struck face of 45 HRC maximum alongside a cutting edge of 53 to 60 HRC, which is the same idea applied to a tool that has to be hit and to cut.
| Specification | Type of tool | Steel composition | Hardness, HRC | Striking test |
|---|---|---|---|---|
| AREMA | Percussion track tools | 9260A, 0.56 to 0.64C, 1.8 to 2.0Si, 0.75 to 1.0Mn. 9260B, 0.51 to 0.60C, 1.8 to 2.2Si, 0.75 to 1.0Mn, 0.35 to 0.50Mo, 0.45Si max | Striking face 51 to 55. Struck face 44 to 48. Cutting edge 56 to 60. Point end 52 to 56 | None |
| ANSI B173.3-1991 | Heavy striking tools | 0.45 to 0.88C, 0.3 to 1.2Mn, 0.35Si max, to ASTM A576, A322 or A681 | Striking face 45 to 60. Wood chopper’s maul 44 to 55. Peen 45 to 60 | Block hardness 92 to 105 HRB |
| BS 876-1981 | Hammers | 0.5 to 0.6C, 0.5 to 0.9Mn, 0.1 to 0.4Si, with 0.2 to 0.3Cr for a head mass above 1.8 kg (4 lb) | Striking face 50 to 58. Three mm below the striking face 46 | Normalized steel block, same steel as the hammer, six blows to the periphery of the face |
| GGG-H-86C-1963 | Hammers | 0.46 to 0.85C, 0.2 to 0.9Mn, 0.1 to 3Si | Double-face hammer 44 to 55. Carpenter’s hammer 50 to 60. Ball-peen hammer 50 to 57 | Handle under 60 cm (25 in.), 92 to 105 HRB. Handle over 60 cm, 40 to 45 HRC |
| ANSI B209.3-1990 | Wood splitting wedges | Carbon steel to ASTM A29, A576 or A681 | 35 max | Drop weight test, 4.5 kg (10 lb) from 1.5 m (5 ft) |
| ANSI B209.1-1991 | Chisels | Carbon or alloy steel to ASTM A29, A322, A331, A576 or A681 | Struck face 45 max. Cutting edge 53 to 60 | Drop weight test |
The six striking and struck tool specifications the source tabulates, with the composition of the steel each one names and the hardness range it allows. All railroad striking and struck tools are currently made from the grade B steel, and nonpercussion tools such as claw bars, rail tongs and rail forks are made from plain carbon steel. Source, Table 1 of the article.
Proving a tool by test
Two of the specifications require the striking tool to be struck against a steel block of 92 to 105 HRB, and a third requires a normalized steel block. Those hardness levels are low, and the source is direct about the consequence. The blocks deform significantly under the blow, and by the same argument used for the struck tool, a soft anvil spreads the stress and reduces the chance that the striking tool spalls. The evidence quoted is that McIntire and Manning were unable to spall hammers with an impact energy of 190 J (140 ft · lbf) struck against an anvil of 24 HRC. A striking test with 100 full swinging blows delivered against such an anvil is therefore a test of the tool’s durability and not of its tendency to spall.
One specification used to require something more severe, an anvil of 40 to 45 HRC for a hammer with a handle longer than 635 mm (25 in.), and that requirement has since been replaced by the ANSI specification with its much softer anvil, with magnetic particle or ultrasonic testing added instead. The other variable in the test is how the blow is delivered. The specification asks for a blow struck squarely with the striking face parallel to the surface, and it asks for glancing blows, overstrikes and understrikes to be avoided, because a blow that lands on an edge puts the load where the section does not support it.
What a spall looks like when it is cut open
A spall that has been examined properly carries more information than the fact that the tool broke. In the ball-peen hammer the source sections, the rim of the cavity appeared sheared and a piece of the striking surface was still attached to it, with the grinding marks on the face ending at the edge of the rim. The cavity surface carried a thin oxide film, which fixed the minimum temperature reached during the event at some point in the 430 to 540 °C range, and the temperature came from the plastic deformation of the spalling process rather than from any heat treatment. In section, an adiabatic shear band covered the entire cavity surface. That band is a light etching layer of untempered martensite, and its profile showed that two separate shears had been involved in removing that spall.
The same examination applied to a different cavity gave the opposite answer. Where the grinding marks run across the rim instead of ending at it, the rim is an indentation rather than a sheared edge, and the cavity formed by ejection of the spall followed by plastic deformation of the surrounding face. Distinguishing those two cases is the point of sectioning a spall, and it is a step that is often unavailable, because striking tools are frequently the subject of litigation and destructive testing is not permitted. Where that happens the spall itself and the fracture surfaces are the only evidence, and the piece that came off is worth keeping for exactly that reason.
| What is seen | What it means |
|---|---|
| A sheared rim, with grinding marks on the striking face ending at the edge of the rim | The material left by the blow was cut off by shear, and the spall came out as a separate piece. Pieces of the striking surface can remain attached to the rim |
| A deformed rim, with the grinding marks continuing across the rim | The rim is an indentation rather than a sheared edge, so the cavity formed by ejection of the spall followed by plastic deformation of the surrounding face |
| A thin oxide film darkening part of the cavity surface | Heat from the plastic deformation that produced the spall. The minimum temperature reached is in the 430 to 540 °C (800 to 1000 °F) range |
| A light etching band covering the cavity surface in a polished and etched section | An adiabatic shear band, which is a layer of untempered martensite. The profile of the band shows how many separate shears were involved in removing the spall |
| A prior crack whose surface is bright inside the part and dark where it meets a free surface | A quench crack. The part of the crack that never intersected the surface stayed unoxidized, and the part that did reach the surface oxidized during tempering, so old and new fracture surfaces differ in contrast |
| Many tools from one lot failing with essentially the same chip geometry | Temper embrittlement, which is confirmed by revealing the embrittled grain boundaries with an etchant such as Rucker’s reagent |
The features a spall cavity carries and the reading each one supports, from the sections and case histories in the article Spalling from Impact Events.
What a large spall says about the steel
The ordinary surface spall is a service event, and a large one is not. The formation of large spalls is invariably associated with a defective material, and the two defects the source names are a prior crack and a change in the grain boundaries. A quench crack is the more common of the two. When a crack of that kind sits entirely inside the tool, its surface never sees the furnace and stays bright, while the part of it that later reaches a free surface oxidizes during tempering, so the old and new fracture surfaces differ in contrast and the boundary of the original crack can be read as a distinct shape inside the cavity. Where the crack was open to the surface from the start, the whole crack surface is oxidized, and the piece that broke away is then much easier to identify as an old defect.
Temper embrittlement is the second cause and it behaves differently, because it leaves no crack to find. The case the source gives is three ripping hammers bought in a single boxed lot that failed in much the same way, with essentially the same chip geometry in each one. The evidence sat in the grain boundaries rather than in the fracture appearance, and it was revealed by etching with Rucker’s reagent. A batch of tools that fails alike is therefore a signal to look at the heat treatment of the lot rather than at the way any one tool was used, and the embrittlement types that can produce that pattern are collected in types of embrittlement in steels.
What this means on a tool steel order
A striking tool order is one of the few where the hardness figure and the failure mode have to be decided together. The hardness window is narrow, the lower end of it is chosen for the warning it gives rather than for the performance it delivers, and the struck tool has to be softer than the striking tool for the pair to work. Three items are worth writing down. The hardness of the face and of the point three millimetres below it, since a shallow hard case over a soft core behaves differently from a through hardened tool. The impact toughness behind that hardness, because a face at the top of the range needs the toughness to survive the blows it will not deform under. And the heat treatment lot, because a batch of tools that fails alike points at tempering rather than at service. Tough shock resisting grades and their heat treatment are on S7 tool steel heat treatment and under tool steels for cracking resistance, the impact energy those grades absorb is recorded in Charpy impact testing of tool steel, and the related wear side of the same impact is on impact wear and spalling of tooling. The way a fracture is classified before it is explained is on the damage mode identification chart, the marks on the surface itself are read with fracture surface features and what they mean, the crack that was there before service is described in quench cracking in tool steel, and the reference for which steel is used for which tool is on tool and die materials by tool type. The wider failure sequence from design to service is set out in tool and die failure analysis.
Before you act on a spall
A reference page, it is not an Aobo Steel specification, and the hardness ranges above are the values published in the specifications named, which apply to the tools those specifications cover. A tool outside those standards, or a tool with a repaired face, is not governed by these numbers. Do not send a striking tool back into service at a hardness above the range because it wore too quickly, since the failure that follows is not gradual. Where a tool has already spalled, keep the piece that came off, because the fracture surface on it may be the only evidence left once destructive testing is ruled out.
Source: ASM Handbook, Volume 11, Failure Analysis and Prevention, ASM International, 2002.