I recently read elsewhere concerning quenching steel in oil. 1095 was the alloy given as example. Should I be changing my procedure? Am I missing something huge here?
Two knives identical. Heat treat identical. Knife A goes into Parks 50 and stays there until it is cooled down to ambient temperature. Knife B goes into Parks 50 for a slow 10 count, and then taken out to air cool to ambient.
What was stated was that knife A would be harder than knife B. Now I THOUGHT that once the PN was reached in proper time, the hardness will not be affected by either cooling it slower in air, or allowing it to remain in the quenchant until ambient. Both knives will reach max hardness....but the knife that was allowed to air cool once the PN was reached might not have as much strain.
NOT that there would be much "strain" at all here. My immediate concern is NOT the relative difference in "stress" caused by the difference in cooling to Ms and down to Mf. But the statement that to reach max hardness, it needs to be left in the oil.
I take all of my knives out of their quenchant after a slow 10 count. Should I be leaving them in the oil?
Thank you!
Marquenching/martempering has been around for, oh… around a century or so, in industry, but knifemakers still struggle with the concept and there is a lot of misunderstanding about what happens. You could take the other sources word for it, or you could take my word for it, I strongly advise you do neither but instead rely on solid information and verifiable data on the topic. I have been playing with marquenching for around 20 years now and have also been testing and slicing up steel for metallography work on the products of that process for about as long. If done correctly there is not hardness loss in the martensite formed, just different HRC readings due to auto-tempering.
Under the microscope I could show this clearly. An older text used the terms “alpha martensite” and “beta martensite” and I always liked the differentiation the terms lend to the discussion. Alpha martensite is as-quenched, it is body centered tetragonal in nature as it retains all the dissolved carbon of the parent austenite; it is typically from 65HRC to 67HRC in hardness. Beta martensite is alpha martensite that has been tempered, it is more body centered cubic in nature and has less carbon in solution than the parent austenite because the process of tempering has formed ultra-fine tempering carbides from the excess solution. The accumulative effect of the tempering carbides is a darkening of the martensitic packets (be it laths or plates, but mostly plates for this discussion); so beta martensite is darker under the microscope and is naturally softer because it has been tempered, so its hardness will be a range that depends in the tempering temperature.
Steel that is quenched all the way to ambient in the oil will be all alpha martensite and so will show the maximum as-quenched hardness. Steel that is allowed to air cool from Ms will also form alpha martensite but as much as 30-40% of the alpha martensite will form at temperatures well within the tempering range and will have time for those temperatures to begin the tempering process. So under the microscope you will see both alpha and beta martensite in marquenched steel. Both steels made the same phase, alpha martensite, just one has had the time to convert some to beta martensite and get a jump start on the tempering process. So while one will typically read about 1.5 to 2 points lower in the as-quenched Rockwell, this is only because one hasn’t seen any tempering yet and will join its partner as soon as it does. So in actuality there is no loss of as quenched hardness just a difference in degrees of subsequent tempering. If the quench was sufficient to get past the pearlite transformation to Ms at the appropriate rate there will be no other phases from the parent austenite other than the martensite, upper bainite can form in minutes but lower bainite takes much greater time and simple air cooling is too quick, so a proper marquench will not involve any austempering processes.
But there are other differences worth noting. Plate martensites impinge its packets at high angle orientations and can lead to embrittlement and micro-fractures of the martensitic plates, and while this can be alleviated somewhat by limiting grain size the effect of carbon over .6% on martensite is still there. The martensitic transformation is shear driven, not diffusional like other phases, so it involves tremendous strain on the steel’s crystalline lattice, the more violent this transformation, the more stress will be involved, increasing brittle type behavior. The slower rate of cooling from Ms not only reduces the stress issues but it also allows for the elimination of some of the strain related problems by a quicker conversion (at least partially) from the body centered tetragonal morphology. This results in less distortion and cracking issues but can also lead to an increase in subsequent impact toughness. My friend Tim Zowada was using marquenching even before I was and took time take samples to a lab for more thorough impact testing, and found he gained around 20% impact strength from the marquenching of O-1 over that quenched to ambient. I have also read research that cites even higher toughness numbers from the process but I am more cautious with numbers over 20% even when provided from a proper scientific study.
So after all that I can say- no, you lose no hardness from the marquenching method, you just gain some initial tempering effects, but you do stand to gain some impact toughness from the process over traditionally quenched steel.