Spark Testing Tool Steel Grades and Chemistry
A spark test grinds a sample against a high-speed wheel until the chips come off burning. Each grade of tool steel burns with its own burst, stream, colour and shape, and an experienced tester compares what he sees against known samples and names the grade. It costs one grinder and a few reference bars, and it is the fastest identity check a tool room has. This page sets out the set-up that makes the reading repeatable, what the burst tells you about carbon and hardness, and what each alloying element does to the spark. It is written for the point where a bar arrives without papers, or a part comes back from service with its history lost.
What the test can and cannot decide
A spark test separates grades and flags a mismatch. Sparking a bar against a reference sample of the grade on the order tells you within a minute whether the two are the same material, and sparking a mixed pile of cut-offs sorts it back into families. What the test does not do is produce a chemistry. Two testers looking at the same burst in different light will describe it differently, and the reading moves with the eyesight of the man doing it, so the result is a match or a mismatch against a sample and never a carbon or chromium figure.
That is the division of labour on a tool steel order. The spark test belongs to the shop floor and to incoming inspection, where the question is whether this bar is the grade that was ordered. The composition on the certificate is a separate chain that starts with the ladle analysis at the mill, as set out on the testing the chemistry of a tool steel heat page, and the paperwork that arrives with the delivery is checked against the order on the tool steel quality verification page.
The set-up that makes the reading repeatable
The spark is a chip of steel burning in air, so anything that changes the chip changes the reading. The published procedure fixes the wheel, the pressure and the light, and it asks the tester to keep a set of samples whose grade and hardness are known. The points below are the ones that decide whether two readings taken a week apart can be compared at all.
| Set-up point | What the test calls for | Why it matters to the reading |
|---|---|---|
| Wheel speed | A high-speed grinder running 10,000 to 20,000 rpm | The spark is the chip burning in air. Below that speed the chips come off too cool to burn in a readable stream. |
| Wheel size and grit | An abrasive wheel 2 to 3 in. (50.8 to 76.2 mm) in diameter, approximately 40 grit | A small wheel and a coarse grit take a light cut that throws a clean stream instead of heating the sample. |
| Wheel condition | Dress the wheel so it is not loaded with grit from earlier grinding | Loaded grit cuts cooler and throws sparks that do not belong to the steel in front of you. |
| Contact pressure | Do not use pressure that slows the abrasive wheel | A wheel dragged down to a lower speed burns the steel differently and moves the burst pattern away from the reference sample. |
| Sample surface | Do not test a decarburized or scaled surface layer | The surface layer is not the steel that was ordered. A decarburized skin sparks like a lower-carbon material than the bar underneath it. |
| Light | Diffused light, a room that is neither brightly lit nor dark | Colour is half the reading and it is read against the surroundings. The same burst reads differently in a bright bay and in a dark corner. |
| Practice samples | Build a set of samples identified by grade and hardness and spark them daily | The bright and dark portions of a burst are learned by eye. A tester who has not looked at a known sample recently is reading the comparison from memory. |
The practice recommended for spark testing, from the spark testing section of Tool and Die Making Troubleshooter. Reference samples are as identified on the samples themselves.
Carbon and hardness in the burst
Carbon sets the size of the burst. A low-carbon steel throws a short, sparse burst and the stream carries on past it, while a 1% carbon tool steel throws a large, complicated burst with many forks. Hardness sits on top of that reading. A hardened bar throws a longer and denser pattern than the same grade in the annealed condition, so a tester comparing a hardened part against an annealed sample has to expect the difference before he calls it a mismatch. That is why a reference set is kept in the same condition as the parts being checked, and why the test is not run on a decarburized or scaled surface, which sparks like a lower-carbon steel than the material beneath it.
What each alloying element does to the spark
Alloying elements work in the other direction from carbon. They suppress the burst and add colour and shape, and the tester reads the grade from that combination. In the tool steel groups the useful separations are the ones between the cold work, hot work and high-speed families, which differ in chromium, tungsten, molybdenum and vanadium content.
| Element | What it does to the spark | What the tester is reading |
|---|---|---|
| Carbon | The higher the carbon, the more plentiful and the more complicated the burst | Burst intensity is the carbon reading. Hardened steel throws a longer and denser pattern than the same grade in the annealed condition. |
| Silicon | Suppresses the carbon burst and burns dark red | A suppressed burst over a dark red stream points at silicon as a deoxidizer rather than at a low-carbon steel. |
| Chromium | Suppresses the spark stream and burns bright orange | The orange colour is the carbides. Chromium raises hardenability and wear resistance and it raises the hardening temperature, which is why the high-chromium families read unlike carbon steel. |
| Nickel | Suppresses the stream and the burst slightly and ends the stream with a forked tongue | A forked tongue at the end of an otherwise quiet stream is the nickel signature. |
| Tungsten | Suppresses the effects of the other elements and burns with a bright orange tongue pattern | Tungsten overrides the rest of the reading, so a high-speed steel sparks in a pattern set mainly by its tungsten content. |
| Vanadium | Brightens the spark stream as a whole | Vanadium is read as overall brightness rather than as a separate burst or colour. |
| Molybdenum | Produces a characteristic spear point | The spear point is the most useful single signature in the tool steel groups, because the hot work and high-speed grades are separated by their molybdenum content. |
| Manganese | Has a minimal effect where the steel carries moderate or large amounts of other alloys | Manganese is a deoxidizer and a hardenability addition, and it does not decide the reading in an alloyed tool steel. |
The spark signature of each alloying element, from the spark testing section of Tool and Die Making Troubleshooter, itself drawn from Bethlehem Steel and Carpenter Technology spark charts.
Read as a group, those signatures are what makes a high-speed steel look different from a cold work steel at the wheel. A tungsten grade buries the signatures of everything else in a bright orange tongue, a molybdenum grade shows a spear point, and a chromium grade shows an intense orange stream with a suppressed burst. Grades and their composition ranges are listed on the tool steel composition chart, and the same elements are what separate one family from another on the tool steel properties chart.
Where the test stops and the certificate starts
A spark test answers a shop floor question quickly, and it is worth being clear about which questions it does not answer. It cannot separate a conventional melt from an ESR or VAR remelt of the same grade, because remelting changes cleanliness and structure rather than chemistry. It cannot measure hardness, which belongs to the hardness testing methods page, and it cannot find internal defects such as inclusions, porosity or centre segregation, which is the subject of the wrought bar defects page and the macroetch and carbide segregation page.
The practical rule is that a spark test is a screen and a certificate is the record. If the screen says the bar is the wrong grade, stop and raise it with the supplier before machining, because a mix-up caught at incoming inspection costs a telephone call and the same mix-up caught after hardening costs the part. If the screen says the bar matches, the mill certificate and the heat number are what the job is documented on.
Where to go next
Grades are described one by one across the grade pages, including D2, A2, O1, S7 and H13, and the application side is indexed on the tool steel material finder. Where the reading is a colour rather than a burst, the temperatures behind those colours are tabulated on the steel heat colour chart.
Spark testing tool steel grades and chemistry, printable PDF
The set-up that makes a spark test repeatable, what carbon and hardness do to the burst, the spark signature of each alloying element, and the point where the test hands over to the certificate, in one reference sheet with our contact details.
Reading this page against your own practice
This page is a reference summary of published shop practice and it is not an Aobo Steel specification or a test procedure of ours. Spark appearance depends on the wheel, the light and the experience of the tester, and a reading is a comparison against a known sample rather than a measurement. Release of material for production is decided by the material test certificate and the inspection practice in force at your plant.
Source, Tool and Die Making Troubleshooter, R. M. Leed, Society of Manufacturing Engineers, 2003.