Notes on Workshop Techniques




Hardening and tempering of steels (and I guess annealing should be included here) is a procedure for modifying the metal's characteristics to better suit the job it has to perform. In the case of toolbits it enables them to machine other materials and still retain a sharp cutting edge, whilst in the case of workpieces it can materially alter their wear characteristics and tensile strength. Virtually all steel hardening processes (as applied by the home machinist) involve the inclusion of particular carbon compounds either at the metal's surface or throughout the body of the metal via the application of heat. It's true that metals can also be 'work-hardened' or forged or laminated but the application of these process (which involve changes in grain or crystalline structure) will not be considered here. Some types of steel (certainly all tool steels) have other trace elements such as cobalt and tungsten in addition to carbon which augment the ultimate hardness achieved and the temperature at which they can operate. The hardening, tempering and annealing of such tool steels ('high-speed' steels as they are known) are beyond the scope of the average home workshop and these steels are usually purchased as shaped bits in the ready hardened state.

At a critical temperature (about 800 degrees C) steel has an affinity for carbon which will be actively absorbed to form a compound called iron carbide. It's an interesting observation that at the critical temperature where the carbide compound forms magnetic attraction disappears, a useful method for determining the correct temperature for hardening purposes. If the steel is then left to cool slowly this carbon-iron compound will spontaneously dissociate back into it's component elements of carbon and iron again. In contrast, if the steel is cooled very rapidly this dissociation does not have chance to take place and the carbon remains trapped in the form of the iron carbide compound. This carbon may already be present in the steel in a variable proportion as part of it's manufacture (but usually not as the carbide form) in which case the steel is known as a 'carbon steel'. Alternatively, extra carbon may be made available to the steel by packing it together with carbon-rich compounds and applying heat over a period of hours, whereupon the carbon is drawn into the outer layers of the metal's surface (a process called case-hardening). Iron carbide is a very hard material but is also very brittle and has little structural strength. The art of hardening and tempering is to balance the very hard characteristics of iron carbide with the toughness, resilience and ductility (i.e., resistance to breakage) of the base metal. An added complication is changes which occur in the metal's crystalline grain structure which is also dependent upon rate of heating and cooling.

After hardening right out - i.e., where the majority of the carbon has been converted to it's carbide form - the metal will be in a glass-hard and brittle form, and is essentially useless because it's mechanical strength is very low. At this point it is necessary to 'draw the temper' whereby part of the iron-carbide is returned back to it's component parts in a controlled manner using precise application of heat. This will soften the metal a little but will also improve it's mechanical strength. The two factors are always in balance and are a compromise condition.

15.1 Silver Steel (Drill rod)

Silver steel is a commercial low-carbon tool steel produced as accurately ground lengths of round and square bar. Common lengths are 13" and 18" (though any reasonable length can be had from the factory), and diameters up to about 1". It is sold in the semi-soft or 'annealed' state. The steel comprises 1.1/1.2% carbon, together with the addition of about 0.35% manganese, 0.45% chromium and 0.1-0.25% silicon. It is particularly useful in the home workshop because it can be easily machined prior to hardening, the latter process being quite simple to carry out with nothing more than a small propane torch and a bucket of water.

Lets take for example a simple 1/2" diameter D-bit reamer, to be made from silver steel and hardened & tempered. The basic shape of the tool would be produced whilst the steel is still in the soft state, being machined or filed to exact dimension. Round silver steel is ground to close tolerance (-0.0000" to +0.0005" on nominal size) so a piece of 1/2" dia rod can be used and simply be cut to length. The metal surfaces should be polished either with fine emery or on a buffing wheel to remove scratches which may later form weak points leading to fracture (not too important for our D-bit but it would be for a home-made tap). The work is then heated to 'cherry' red (somewhat above the transition temperature of 800 degrees) to form the iron carbide before quenching in cold clean water. Brine is actually a little better as the bubbles formed are finer. Cherry red is a term often used but in fact is quite difficult estimate, being dependent as it is upon the levels of ambient light. Perhaps a better method is to use the disappearance of magnetic attraction as a measure of the critical temperature. Quenching should be done by dipping the hot metal bar vertically into the water then swirling it around a little, this technique will minimise any tendency to warp which might occur should one side cool (and therefore shrink) appreciably faster than the other.

Upon removal of the D-bit from the quench it will be noted that the surface is black and probably scaled. An old-time tip to avoid this is to coat the metal with soft soap prior to heating, upon quenching the metal will retain a nice even silvery-grey surface. In any case, the metal now needs to be polished back to base metal to facilitate the tempering process. One characteristic of polished steel is that upon heating through a range 150 to just above 300 degrees C it progressively changes colour from faint yellow through straw, brown, blue and finally near black. We can use this colour change to indicate the temperature of the metal and so control the drawing of the temper. Each stage of the colour change indicates a percentage of the carbide compound remaining in the steel, and therefore it's hardness. The objective of tempering is to reduce the hardness to the point required and then stop the carbide breaking down any further by immediately quenching in cold water. Useful indicators are:

You can see from the above that there is considerable range in the actual hardness achieved, and this is due variations in temperature (for which colour is only a relatively crude indicator) and also variations in the carbon content of the parent steel. Nevertheless, it's clear that a home-made tap tempered to pale-yellow might be useful for tapping hard materials but would be very likely to snap in use due to it's low tensile strength. Our D-bit is unlikely to be subject to the same tensile stresses the tap would be so we can get away with somewhere between pale-yellow and pale-straw to take advantage of the extra hardness. Note that these temperature are relatively low and, for a lathe toolbit, or home made milling cutter destined for arduous work, such temperatures might be achieved in use. It is clear then that hardened low carbon steels must be kept cool if they are to retain their hardness, a property which is the main reason why 'high-speed' steels have displaced them for normal machine tool usage.

The appearance of 5 pieces silver steel hardened and tempered as described in the table above.

The D-bit example is fairly straight-forward and is a good place to try out your metal hardening skills. However, there are many jobs that require differential tempering such that one part of the tool is softened and another stays hard. It is here that some skill comes in to play. Imagine a simple metal rod some 1/4" diameter, it's possible to heat this rod with a small flame some 2-3" from the end, and watched carefully the colours will be seen to 'run' towards the end. It is quite possible at this point to remove the source of heat and watch the play of colours and just when the correct colour is achieved at the tip quickly quench. This will give a range of hardness that ranges from perhaps very hard at the tip to soft some 2 inches further along. You can control this process even more using 'heat sinks' which will slow the rise in temperature at one point but not at another. A favourite trick is to sink the end which needs to remain hard in a potatoe which will then keep that portion relatively cool. Other workpieces may be of such fine structure that it is difficult to apply the heat evenly. In this a case it is necessary to place the object in a sand tray and heat the lot from below, a larger piece of steel is usually included in the sand bath to act as a visible indicator of the temperature attained, and once attained the whole lot is dropped into the quench bucket. This method is particulaly used for small metal parts (like watch hands) to produce a nice even blue colour, and also for springs to prevent the formation of hard spots which might lead to subsequent fractures.

Return to top

15.2 Gauge Plate

Gauge-plate (or ground flat stock) is an oil-hardening carbon tool steel which can attain a hardness of about 880 Vickers. Oil is used as the quench because it is somewhat slower than using water, and there is little formation of bubbles. The tendancy to warp with an oil quench is less than that with water but it is still very important that the workpiece should be immersed in the quench along it's critical axis (i.e., in the case of a 3" x 1/2" x 1/4" plate which needs to remain straight along it's major dimension it would be inserted lengthways into the oil). Tempering is performed much as with silver steel, and colour change can be used to estimate hardness. A more useful method is to to immerse in a sand bath and use a domestic oven to attain the correct temperature (240 degrees C for gauges, and 150 degrees C for cutting tools).

A few words on annealing may be appropriate here. Sometimes it is necessary to soften tool steels which have been hardened by the previously described processes. Commercial items such a old files can be used for all sorts of things if only they could be softened first so that they can be machined. This is quite a reasonable proposition but it requires some care. It is most important that the cooling process after heating to the transition point be very slow, if this is not so then upon re-hardening it is very likely that the metal will fracture in use. In the case of the old file it can be heated in a coal fire which is then left to burn out overnight. The metal can be machined and then re-hardened by heating, quenching and tempering again. Another example is the common necessity to anneal work-hardened sheet copper in the process of making a model locomotive boiler. In this case what was happened is that distortion of the metal has caused a coarsening of the grain structure which if not addressed will eventually cause the metal to fracture. To correct this situation it is necessary to heat the metal to bright red heat and rapidly quench (i.e., the self-same process used to harden carbon steel). However, in the case of soft metals like copper what happens is that the crystalline structure is returned to a very fine grain, increasing ductility and enabling it to be bent easily once more. It can be confusing that the same process that hardens one metal softens another but it becomes clear when you know the reasons.

Return to top

15.3 Case Hardening

The process of case hardening is particularly useful for workpieces made from mild steel which are subject to wear and would therefore benefit from a hardened surface. Unlike carbon tool steels the hardness attained only applies to the outer surface of the metal (between 1 and about 15 thousandths of an inch deep) so the inner core of the metal retains it's original ductile properties. For items which require only a very thin case (simple plain bearings or shafts) it is possible to use a commercial product called 'Kasenite'. This is an easy and rapid process where the item to be treated is heated to bright red heat, dipped in the hardeing material, and heated again whereupon the material melts and coats the metal surface. The heat is maintained for a few minutes and then the item is quenched in clean cold water. A somewhat thicker case can be obtained by repeating the process but it is a case (no pun intended) of diminishing returns - each application increases the depth by a smaller amount. Probably the maximum obtainable is about 0.003".

A thicker case requires that the item be packed in a metal tube surrounded by carbonacious materials (bonemeal, rags etc.,) and sealed before placing in a furnace. The item is then heated for several hours to allow the carbon content in the surface of the metal to build up. This method can give a very thick case and is usefully applied to machine tools which are subject to considerable wear and tear.

Case hardening can also be used to make single-use form tools (for cutting metal) where it would be either impracticable or too expensive to use ground flat stock - which is relatively expensive. Certainly for use on brass and aluminium, and for limited use on mild steel, such a tool can be easily machined and hardened.

Such items as pivot pins, with male threads at each end, are candidates for case hardening. If a plain metal rod is first case-hardened to carbourise (the correct terminology for case-hardening) the surface, then annealed, it can still be easily machined (just like silver steel can). The threads can be machined onto the ends of the rod and then the item heated and quenched. This will leave the threads un-hardened (and therefore unlikely to break off) yet will leave the main body of the pivot hard. Such differential hardening simply requires a little planning prior to the machining stage.

Return to top

(c) Chris Heapy 1996.

Back to previous page