As promised I decided to start a new, hopefully educational, thread just dealing with the intricacies in complex mechanism involved in quenching steel. I have recommended the book Quenching and Martempering by ASM, but until those interested can get their own copy I will touch on some of what is in that book for starters.
Why do we quench? This may sound like a stupid question due to what we perceive as the obvious answer, but it is not that simple. Aside for other reason for cooling, I will assume we are hoping to harden the steel with this operation. Hardening the steel is a little subjective, as any state at all harder than the previous is technically hardened. So if we heat fully annealed steel and then cool it in air to form very fine pearlite the steel will indeed be harder than before. If we cool the same heated steel in a medium that will keep it from 500F-800F for extended periods, upper or lower bainite would in fact be harder that the previous condition. The same steel quenched to from martensite will be the hardest yet, but severity of quench below 500F could increase strain and result in even more levels of hardness. Then we need to discuss types of hardness, an air cooled piece of steel with subsequent cold work will exhibit high levels of strain hardening, which will resist bending, but not scratching and cutting. Mixed microstructures and carbide distributions will have great scratch hardness by may move significantly under penetrative hardness readings.
So before we can decide what quenching is even all about, we need to determine what hardening is all about for us. For simplicities sake, and since the highest levels of strength and abrasion resistance, in the absence of carbides, come from the most homogenous martensitic conditions, I will confine the scope of my discussion to achieving the most martensite formation, without undo brittleness, as possible. Do we all agree there?
Making martensite can actually be simplified down to not making anything else. With the starting point of austenite, cooling steel will want to form phase more stable than that at a give n temperature. The greatest challenge is pearlite, from 1,000F down to perhaps 900F pearlite will very rapidly form from the parent austenite, thus the most critical function of quenching simple steels is cooling as rapidly as possible through this range. The next phase that can form in the range from 900F to 500F with higher carbon steels would be the upper and lower morphologies of bainite. Some may choose lower bainite as their goal but to keep the topic manageable I will not get into austempering. All these other phases at least partially involve diffusive processes, the one we seek must avoid diffusion of the carbon in order to achieve the condition necessary for its formation. Thus if we want to make martensite the goal is to cool austenite at such a rate so as to outpace any diffusive activities that could rob us the carbon necessary for the strain of martensitic transformation.
With a steel such as 10XX series, there is not drag form alloying to slow down diffusion and we must cool entirely through the pearlite range in less than one second. This speed will obviously be heavily effected by the thickness of the steel, so the thicker or blade the faster the quench needs to be. Thus in choosing a quenchant our first reaction will be to go with the fastest cooling rate possible however this is where things begin to get complicated and such an oversimplified view will cause serious problems. As one approaches 500F and below the need for speed grows less critical and can take a back seat to another concern- over stressing the steel as it undergoes the incredible strain of the martensite transformation. Here continuous cooling is important in order to maintain that transformation but cooling at the same rate or faster, than that needed to avoid pearlite could result in distortion or cracking. So the ideal quenchant for most steels will cool very fast in the initial phases and then at a slower rate in the final 400 degrees. If this were not the case then brine would simply the best all around quenchant, and yet it is not.
The phases of a liquid quench: When hot steel in introduced into most liquids used for quenching there will be at least four distinct phases on its reaction to the steel being cooled to ambient- Vapor formation, Vapor blanketing, Vapor discharge, and final conductive cooling.
Vapor formation occurs as soon as the hot steel superheats the liquid to instantly exceed its vapor temperature, with water it will be 212F, with many oils it will be between 375F and 450F. This is almost unavoidable with almost any liquid with a vapor point below the temperature of the steel.
Vapor blanketing is what happens when the steel superheats enough liquid to create a constant and solid blanket of insulating gas which can all but stop the conductive cooling effects of the quenchant. Unfortunately this effect occurs in the same range when it is most critical to cool the steel in order to avoid pearlite. To add to this the vapor jacket will tend to perpetuate itself by its insulative effects on the steel. Thus not only is one of the most critical attributes of a good quenchant a very low vapor jacket, anything that can be done to destabilize or reduce that jacket is highly important in a successful quench.
Vapor discharge is the most violent and rapid cooling step of the process. This is when the vapor jacket collapses and the liquid begins to make direct contact with the hot steel, resulting in violent boiling and even some small gaseous explosions. Put a hot piece is steel into water and there will be an initial hiss from the vapor jacket formation, followed by a violent vibrating rumble as the water bites the steel in the vapor blanket collapse. This phase also self perpetuates as the boiling action increases convective forces of the quenchant as it self agitates. The more evenly this jacket dissipation occurs the better the chances of avoiding distortion and uneven hardening.
Direct conductive cooling can finally occur once the temperature has lowered enough to avoid heavy vapor formation and the liquid itself is making direct contact with the steel and carrying the heat away with conduction aided by convection of the heated liquid. Due to both of these factors working unhindered this has a natural tendency to be much faster that may be desired.
Many people have experiences problems with water quenching blade steels, let us examine how water, the oldest of quench mediums, behaves in these stages. First it forms vapor at a much lower temperature than oils and in voluminous amounts, so it will readily create a very insulative jacket of steam surrounding the steel, the vapor discharge will be most violent and very uneven, resulting in much greater distortion and varied rates of cooling. Once the vapor jacket is entirely gone the rate of cooling will be very fast to ambient, increasing the problems that were started in the previous step and often subjecting those uneven forces to enough stress to result in fracturing. Adding brine to water does not do much for us in the way of the actual conductivity of the quenchant, but what it does do is destabilize the vapor jacket to the point that the cooling can begin much sooner and much more evenly.
Oil can have greater or lesser degrees of vapor jacket problems depending on the vapor point of the oil. In general oils offer less vapor interference but a lower thermal extraction rate, due to limitations in conductivity and convection. Less viscous oils have a greater ability to convect and thus move heat ways from the steel and this is one of main reasons slightly heating a quench oil can increase its cooling ability. Perhaps the greatest advantage of oils is the gentle transition from the vapor jacket phase and the slower direct cooling phase, resulting in less distortion or cracking during martensite formation.
To be continued
