7 Heat Treatment of Steel
PART ONE
Hardening
We are here dealing with the kind of heat treatment normally done in the small workshop for which we need only use the brazing hearth and torch.
More information regarding the structure of steel is given in Part II. Although this is kept in simple terms it is probably best understood as a subject if done in conjunction with the science labouratory where there are suitable microscopes and possibly some testing equipment.
For hundreds of years such small tools as chisels, punches, knives, shears etc, have been hardened and tempered by metalworkers using similar methods to those mentioned here.
Many special alloy steels are available but these have to be treated as advised by the manufacturers. We are considering only plain carbon steel.
Steels may be roughly classified as follows: Mild Steels 0.1—0.33% carbon. Medium carbon steels 0.34—0.60% carbon. High carbon steels 0.60—0.90%C and tool steels 0.90%—1.3% carbon.
For small tools we use steel with a carbon content of between 0.8—1-3% carbon. Files fall into this category and for this reason small tools are often made from old files.
If we heat a piece of carbon steel with the brazing torch until it is red hot and then plunge it quickly into cold water we find that it has become "dead" hard. This can be tested with the heel of a file, i.e. the part of the teeth nearest the handle, so as not to spoil the working part of the file.
The primary problem in hardening is judging the correct temperature at which to plunge the steel into water. This temperature is commonly known as "cherry red" and it is best judged in a shady corner away from direct sunlight or bright electric light.
If the steel to be hardened is heated with the torch the temperature will rise slowly until it reaches "cherry red" but then although the torch is kept on the steel it will cease to rise in temperature for a few seconds. This is known as the change point; the heat from the torch on the steel is now latent, that is "hidden", because it is used to bring about changes in the structure of the steel. This taking in of heat is called decal-escence (see fig. i), and it occurs between 700°C and 900°C approximately, depending on the carbon content.
It is at this temperature that we plunge the steel into the water and in so doing we arrest it in its changed state.
A way of showing this change at red heat is to heat the metal well above "cherry red" and then allow it to cool slowly in a darkened corner of a room. As the temperature slowly drops to the change point it will be seen to glow suddenly. This is caused by the latent heat being given out. This is called recalescence (fig. 1). However, this is only for demonstration purposes and it must not be used as a method for arriving at the temperature at which to quench the steel. The best results are obtained if the steel is slowly brought up to the change point and then plunged; that is, it is best plunged on a rising temperature.
The hot metal is usually plunged into clean water which is at room temperature. It is essential that the quenchant (the liquid in which the hot metal is quenched) takes the heat away quickly in order to arrest the steel in its changed state. If the cooling needs to be more drastic, cooking salt can be added to the water to make it into a saturated solution. This is best judged by adding the salt to the water until a slice of potato will float in it. This, however, increases the possibility of cracking. Cracks might occur even when using clean water. If this is happening it is often because the steel is being overheated, but it might be because the quenchant is taking the heat from the steel too rapidly.
If soap is added to clean water it will make it less severe as a quenchant. An even less drastic quenching medium is high flash paraffin and a slower one still is mineral oil.
It will also be noticed that cracks occur on steel where there are deep scratches or small grooves such as those left by a turning tool. To avoid this it is best to remove the rough surface and polish the steel before hardening.
Be careful when quenching long thin tools such as chisels, gravers or knife blades, that they are plunged into the liquid vertically and that they are moved up and down, not from side to side, otherwise distortion might take place. If they are plunged horizontally the side entering the water first will contract first, thus causing the steel to be badly distorted. It is good practice to stir the liquid vigorously just before plunging any piece of steel as this helps the liquid to conduct the heat quickly and evenly.
As previously mentioned cracks occur in steel, particularly large pieces, if they have been taken to a temperature too far above the change point and then quenched. This is owing to the fact that when steel reaches the change point it is also at its state of least density. In other words, when it is at this temperature of change it has expanded to its maximum.*
* To be precise, further expansion docs actually take place at higher temperatures.
If it is quenched above this temperature it undergoes cooling contraction, the skin becoming hard and rigid. The inner portion has not yet felt the quenching effect and is still red hot. An instant later the quenching effect is transferred to this portion, which as it passes through the change point must expand, thus causing cracks on the outside.
For this reason large tools or tools which have a drastic change of section (fig. 2) should be heated to a temperature just approaching "cherry red" and, depending on the carbon content, dipped in oil or high flash paraffin.
Completely hardened steel is usually too hard and brittle for normal use and so we have to reduce the hardness by a secondary process known as tempering.
Tempering
Tempering reduces the hardness of the steel and increases the toughness, i.e. the capacity to withstand shock. The degree of toughness required is determined by the kind of job the tool has to do. Tools for turning brass, scrapers and engraving tools have to be much harder than repousse punches and these in turn must be harder than a screw driver which is just soft enough to be filed and yet tough enough not to be deformed when a twisting force is applied.
To temper a piece of hardened steel, polish the hardened portion with clean dry emery cloth (avoid using oil as this leaves a thin film) and let a small flame play on the metal a short distance from the part to be tempered (fig. 3A). Soon you will see the bright metal near the flame turn light yellow and then straw to middle-straw and then light brown to dark red brown to purple and then to blue. As the heat is conducted along the metal these colours which are in fact oxides will move along in bands. The lightest colour will be the one furthest from the source of heat. When the right colour has reached the part to be tempered, the steel is quenched in clean water. The lighter the colour (best seen in daylight) at which the steel is quenched, the harder the metal will be.
| °C | TEMPERING CHART | |
| Light blue | 315 | Too soft for cutting edges. |
| Blue | 300 | Springs, saws for wood, screwdrivers. |
| 290 | Carving knives, fine saws, saws for bone and ivory. | |
| 285 | Needles, gimlets, axes, adzes, augers. | |
| Purple | 270 | Flat drills for brass, cold chisels for light work, wood borers. |
| Dark straw or red brown | 260 | Wood chisels, plane irons, stone cutting tools, axes. |
| 250 | Flat drills, reamers, taps, screwing dies, shears, punches, chasers. | |
| Middle straw | 240 | Pen-knives, circular cutters for metal, boring cutters. |
| 235 | Milling cutters, lathe tools, wood engraving tools. | |
| Pale straw | 230 | Surgical instruments, razors, hammer faces, ivory cutting tools. |
| Pale yellow | 220 | Steel engraving tools, scrapers, light turning tools. |
If a uniform temper is required along the whole length of the steel it can be held in light tongs in a tube which is heated on the outside (fig. 3B) or held over a heated plate (fig. 3C).
Beginners often have failures with tempering because they misjudge the colour at which to quench. It must be remembered that if in tempering, the steel is heated beyond the proper colour the whole process of hardening and tempering must be repeated. If, however, after quenching the temper colour is not dark enough it will do no harm to re-polish the tool and re-temper until the darker colour is reached then quench.
One Heat Hardening and Tempering
This is the method often used by blacksmiths usually on the end or point of tools.
The steel is heated at one end to cherry red and the tip only is quenched. It is moved up and down slightly to reduce the possibility of cracking at the water line. Then it is taken from the water and the end is briskly rubbed with an old piece of carborundum stone or something similar so a portion of the tip is bright. After a few moments the heat from the unquenched part is conducted to the tip thus showing the usual bands of colour. When the correct colour reaches the tip it is completely quenched.
Case Hardening
This is a method of hardening mild steel by adding carbon to the "skin" or "case" of the steel and then quenching it as if it were a high carbon steel.
The work to be case hardened is put in a shallow steel tray or shallow open box and a powder rich in carbon (usually "kasenit", which is a proprietary brand) is sprinkled on top to cover the work. With a brazing torch the work is made red hot. This causes the carbon to be absorbed by the steel. The longer the work is kept at cherry red the deeper the carbon will go. It should be kept at cherry red in the kasenit for at least 15 minutes and then quickly removed and quenched in clean water. Beware of the loud "bang" it makes. If tested with the heel of a file it will be found to be hard. Usually, in school, because of the time factor it is not possible to make the hard case any more than about 0.007" deep.
Case hardening is excellent for such things as spanners where a hard surface and a tough core are essential.
Softening Steel
To make steel as soft as possible it must be heated to just above the change point and then quickly buried 6" to 8" deep in slaked lime or vermiculite which should first be heated. Slaked lime and vermiculite are very poor conductors of heat and they allow the steel to lose its heat very slowly, thus allowing the change to the soft state to be gradual. This process is known as annealing.
The easier workshop method which makes the steel soft though not quite as soft as by annealing is done without slaked lime or vermiculite. The steel is heated to the change point and it is left to cool on the edge of the hearth or on a fire brick. This is known as normalising.
The purpose of both of these processes is usually to make a hard piece of steel soft enough for it to be machined, filed, sawn, bent or twisted.
Softening relieves the internal stresses. Any tool which has to be re-hardened should first be softened to relieve the stresses set up by the first hardening. This helps to prevent cracking.
PART TWO
We have mentioned the workshop methods in Part 1. In this section we are looking a little deeper into the heat treatment processes and the structure of steel.
Steel is basically an alloy of carbon in iron, although other elements may be present either in residual amounts or as intentional additions to give specific properties.
Carbon in Steel
Carbon forms a chemical compound with iron of the formula Fe3C (containing 6-68% C) which is known as cementite and may appear in steel either individually as the compound or intimately mixed with virtually pure iron to form pearlite. The latter contains 13% of cementite and 87% of the nearly pure iron called ferrite. Under a microscope pearlite is seen to consist of thin plates of ferrite interleaved with thin plates of cementite. Pearlite is so named because under certain conditions it has a pearl-like lustre.
Iron
Iron (Fe) may exist in different physical forms, even though its chemical properties do not alter. This characteristic is known as allotropy, and is common to other elements, such as sulphur.
Metallurgists name the different forms of iron after Greek letters, i.e. alpha α, beta β, gamma γ, and delta δ.
Iron in the cold state is known as a iron and it remains so up to 768°C when it changes to β iron, where only a magnet change takes place. At 910°C it transforms to γ and then at 1400°C to α.
These are changes in the arrangement of the atoms, but the important change so far as the hardening of steel is concerned, is the change from γ, known as austenite, on cooling below 723°C.
The change from a to β occurs at 768°C and above this temperature the iron is non-magnetic.
The change to γ iron at 910°C on heating is accompanied by a marked contraction.
The austenitic (γ) form of iron can hold carbon in solid solution up to approximately 2%, whilst ferrite (α) can retain only about 0-002%.
In steel the existence of carbon up to a maximum content of 0.87%* dissolved in γ iron is able to depress the temperature at which the change from γ to α occurs and the resulting precipitation of cementite commences. Beyond this critical content however, the change temperature rises again. This can be seen more clearly if we plot the level of change-point temperatures against carbon content in steel as shown in figure 4. This diagram shows that as the carbon content increases to 0.87% so the temperature of change drops from 910°C to 723°C. Above the 0.87% carbon content, however, the temperature at which the change begins starts to rise again.
* This figure varies wilh the purity of the iron.
Figure 5 is a similar diagram to figure 4, but the individual points have been joined and the fine so formed is labelled the upper critical line". A horizontal line is also drawn through .he minimum transition temperature reached and this is called the "lower critical line". Thus we have a simple version of what is known as an "equilibrium diagram".
From figure 5 we can see:
1. At all temperatures and carbon contents above the upper critical points the carbon is held in y iron in solid solution, a form of alloy known to metallurgists as austenite.
2. The minimum possible transformation temperature occurs with 0.87%C, the change from y to a iron throws all the carbon out of solution at once to form pearlite.
3. The change from γ to α iron takes place over a wider range of temperature as the carbon content is reduced below 0.87%. This range is indicated as the difference between the upper and lower critical points. The first constituent to appear is ferrite and this goes on appearing until the carbon content of the remaining y iron has been enriched to 0.87 %C, then this changes at once to pearlite.
4. At carbon contents above 0.87% C the change also takes place over a range of temperature, but this time the first constituent to appear is cementite, the compound of iron and carbon of the formula Fe3C. This goes on forming until the carbon content of the remaining austenite is reduced to 0.87% at which time the remainder changes to pearlite.
Now we can complete our simple equilibrium diagram as shown in figure 6.
Carbon Content and its Effect on Hardness, Toughness and Ductility
Ferrite is almost pure iron. It is soft and ductile. Steel containing a lot of fcrritc will have a low tensile strength and will be tough and ductile.
Pearlite is a hard material mainly because of its intricate laminated structure of thin plates of ferrite and the extremely hard plates of cementite. As the carbon content increases up to 0-87% so does the hardness of the metal as a whole, but the toughness and ductility decrease.
Cementite. It is very hard and brittle. Steels containing more than 0-87% carbon have free cementite at the crystal boundaries. This causes extreme hardness in the metal as a whole and care must be taken with the heat treatment. It is used where great hardness is required, such as in ball-bearings, ball races and tools.
The Crystalline Structure of Steel
The crystal structure of steel can be seen under a microscope but the metal must first be prepared. The piece to be examined is first ground flat. Then it is polished with successively finer grades of emery paper and finished on soft cloth impregnated with alumina powder or diamond paste. The polished surface is then etched with dilute acid which attacks the different constituent parts differentially, thus revealing the crystalline structure. The four diagrams shown give an idea of the appearance of steels when seen through a microscope, but of course, no two pieces ever look exactly alike.
1. Ferrite or pure iron is as shown in figure 7.
2. Steel containing less than 0.87% carbon, figure 8.
3. Steel containing 0.87% carbon (eutectoid steel), figure 9.
4. Steel containing more than 0.87% carbon, figure 10.
Rate of Cooling and its Influence on Structure
Annealing. Annealing as previously mentioned is done to produce the maximum softhess in metal. A piece of say 0-5% carbon steel is heated to just above the upper critical temperature and then allowed to cool very slowly in the furnace, i.e. the furnace is switched off and the metal cools at the same slow rate as the furnace.
As the temperature falls ferrite is precipitated first forming new crystals of this nearly carbon-free constituent around the original austenite crystal boundaries. As more ferrite comes out it will form a series of large new crystals. At the same time the remaining austenitc will be enriched in carbon until it contains 0.87% carbon. When the temperature drops through the lower critical point the remaining austenite will change at once to pearlite (fig. 11).
In annealing the slow cooling ensures that the carbon is allowed time to diffuse into the austenite as the fcrritc appears along the crystal boundaries. This gives annealed steel a comparatively coarse crystal structure.
Normalising. Normalising is similar to annealing except that the rate of cooling is faster, the metal being allowed to cool in still air. Draughts may cause it to cool too quickly.
The metal is heated in the same way as for annealing, but because the cooling is quicker the carbon docs not have time to diffuse so easily through the metal and small crystals of ferrite appear both round the crystal boundaries and in the centres of the original crystals as in figure 12. Normalised steel has a finer crystal structure than annealed steel and is slightly harder.
Low carbon steels from which boiler plate and some girders are made are usually normalised.
Quenching. When steel is quenched the cooling is so rapid that carbon does not have time to diffuse through the mass. Instead the metal is "frozen" and the carbon is prevented from precipitating as iron carbide (cementite) below 723°C and will remain in super-saturation until the temperature drops to a value where an instantaneous transformation will take place to form a product known to metallurgists as martensite. This product has a characteristic needle-like structure (fig. 13) which is extremely hard and brittle due to distortion of the ferrite crystal lattice by carbon. This brittleness must be reduced before the steel can be successfully used in service and therefore it is usual to follow the quenching by a tempering treatment.
Tempering. Tempering allows some of the carbon to redistribute itself as iron carbide and this reduces the stresses in the steel.
Only a relatively low temperature is required (up to 650°C) to permit this redistribution to take place, producing a structure of iron carbide finely dispersed in ferrite, in the past known as sorbite, but now more correctly known as tempered martensite. This tempered steel is less hard but much tougher (fig. 14).
Case Hardening. In Part 1 we mentioned case hardening mild steel using "kasenit". A piece of mild steel after being heated to just above the critical range for about one hour in "kasenit" and allowed to cool slowly would appear as in figure 15.
Notice that the body of the mild steel contains a very small amount of carbon, but not enough to affect the hardness.
If the same steel is rapidly cooled from just above the critical temperature by plunging it in water, the outside or case would become martensitic but the core, not containing enough carbon, would be soft. It would appear as in figure 16.
Low carbon steels can be case hardened to great advantage. Because the case contains more carbon than,the core the upper critical point of the case is lower than that of the core. The steel is quenched from the high carburising temperature thus making both the core and the case martensitic. Next it is re-heated to a temperature which will just turn the case (higher carbon content) austenitic but only hot enough to temper the core (lower carbon content). When quenched from this temperature the case will be martensitic and therefore hard and the core sorbitic and therefore tough.
There are various other methods of case hardening such as: sorbitising. This is not unlike the blacksmith's "one heat" hardening and tempering mentioned in Part I. The red hot steel is taken from the rolling mill and the surface sprayed with water thus making it martensitic. The heat from the interior of the steel as it slowly cools then turns the martensite into sorbite thus giving a tough outside and a less hard centre.
For further reading see list of books at back.
HEAT TREATMENT OF NON-FERROUS METALS
In the small workshop we use copper, brass, gilding metal, aluminium, aluminium alloys and, less often, zinc and nickel silver.
All the above mentioned metals work harden, i.e. the crystal structure becomes distorted with bending and hammering. They can be softened in the brazing hearth using the torch for heating.
Copper. To soften heat to dull red and either allow to cool in air or quench in water. Quenching in dilute sulphuric acid will soften and clean the surface (see Chapter 9).
Brass. Soften by heating to dull red and allowing to cool in air. Quenching can cause some brass to crack.
Gilding Metal. Soften as copper.
Aluminium. Soften by heating and quenching in water, but take care not to overheat. The temperature can be judged by rubbing a piece of wood on the surface of the metal. When the wood leaves a black charred streak the aluminium is hot enough to plunge into cold water. This will make it quite soft. The correct heat can also be judged by rubbing streaks of soap on the cold metal. Heat until the soap turns brown and then quench. Another method is to put oil on the surface. Heat until the oil burns, then quench. In schools the first method has been found to be the most reliable.
Aluminium Alloys. Soften as aluminium.
Zinc. This can be made workable by placing in boiling water.
Nickel Silver. Soften as brass. Never quench.
<
