Following much discussion on Facebook recently (Jan 2017) I thought I’d put pen to paper (or at least today’s electronic equivalent) in the hope that I can answer some of the more commonly recurring questions. My main problem is trying not to make this too technical so if you need more clarity please contact me (contact form on this website, my Facebook group “Graham Clarke Knives, etc).
Simply, the quench part of the heat treatment process is cooling a hot piece of metal. This can vary from using iced brine (extremely fast) to burying in vermiculite (very slow). The area I’m going to address in this article is the one of quenching a knife blade to get the desired results in the finished blade.
Starting point :–
- Steel blade between 2.5mm and 8mm thick.
- Blade heated to correct hardening temperature in a furnace
- Yes, I know many of you want to heat in your forge fire till the magnet drops off and I know you’ll get good results because you’ve done it this way for the last 87 years so why change!!! I can only discuss here the most scientifically correct way to do it.
- So now it’s at the correct hardening temperature and next comes the quench. What are we trying to do next?
- If it’s already hard and you’ve forgotten to drill a hole bury it in vermiculite so that it cools very slowly and it should be soft enough to drill. This is very unlikely to succeed for a stainless steel as it will still not cool slowly enough so for these just heat to 6300C for an hour and air cool. You may still find it difficult to drill but this is the best you’re going to get in a knifemaking workshop situation.
- You want to harden it (and keep it straight?) and temper it to make a blade that’s usable as a knife. So quench it – in Brine? Water? Caustic soda? Oil? (fast? medium? slow?), polymer? etc with or without agitation? and then remove it from the quenchant at what temp?? and then freeze it? (before temper or after?) or not? – fridge? freezer? dry ice? liquid nitrogen??????
The choices are mind boggling so this is where I’m hoping to debunk a few myths.
- You must quench into a recognised quench medium. What you’re currently using may not be a recognised quenchant and may be working for you.However, you probably won’t be giving yourself the best chance for top results. Recognised quench mediums are:-
- Water – with or without additives such as salt, caustic soda or special quench polymers.
- Oil – correctly formulated quench oil with the correct additives to give the required quench speed. The speed at which your quenchant cools your blade is THE most important factor in the process. Transformer oil, mixtures of ATF and diesel, old engine oil will all give a hard blade but probably with soft or softer spots and almost certainly will not achieve optimum hardness.
- Air – Still, forced, enclosed in the hot furnace (furnace cool), trapped in vermiculite etc.
- You must be quenching from the correct temperature. This does not mean that you cannot quench from your forge fire when magnetism is lost but you’ll be less likely to be at the optimum temperature than if you are using a controlled temperature furnace. Stainless steels are much more sensitive to incorrect hardening temperature than carbon steels.
- You must allow the quench process to proceed through the correct stages (some of which may only be a second or two). Seeing Bladesmiths on “Forged in Fire” removing carbon steel blades from the oil and getting a massive flare makes me cringe, yet I do something similar with my stainless blades as part of my process to ensure they’re fully hard and straight!!!
What follows only applies to blades made from proprietary carbon or stainless steels. Pattern welded steels may require some variation (it’s NOT a matter of calculating average carbon content – that’s just wrong). Please contact me if you have a specific metal combination you’d like advice on.
Martensitic stainless steels that we use for blades (12C27, 14C28N, 440B or C, N690, the PM steels etc) will all harden with an air quench or slow to medium oil. I have cracked a few with fast oil!!! This intrinsic good hardenability gives us some options that we simply do not get with carbon steels. We need to understand what happens during the quench though to take full advantage of this characteristic.
When any steel starts cooling from the hardening temperature there is a critical time/temperature point to which it MUST cool within a SPECIFIC TIME PERIOD. If you’re familiar with TTT or CCC diagrams this is known “getting past the knee” of the cooling curve. To explain, below is a TTT (Time Temperature Transformation) diagram for 440C.
Temperature is on the vertical scale and time horizontal. Note time is on a logarithmic scale as all the action happens very quickly at the start and then slows down dramatically. The curve on the diagram that appears between 7000C and 6000C is the knee. I have drawn a typical cooling curve representative on an oil quench in blue and an air quench in red. Both lines “miss” cutting the knee. Then, as the line cuts the lower black line (Ms) the metal starts to transform from the Austenitic (soft_ phase to the Martensitic (hard phase). That’s why the line is marked as the Ms (martensitic start) line
However, it is important to note this phenomenon of the steel converting from soft to hard after it has cut the lower (Ms) line. Therefore, if we were to try to straighten a warped blade before it gets to, say, 2500C it will still be soft, can be straightened, will stay straight, and will not move again as it hardens. I use this feature to straighten my blades by getting them into a press between 2 flat plates as they cool. I achieve this by removing the blade from the oil at about 4000C (oil auto ignites at about 4500C) so as long as they don’t flash but are smoking heavily when I remove them from the oil I’m at about the right temp. Time in the oil varies from about 6-7 seconds at 2.5mm thick to 10-15 secs at 4mm and above. The green line shows that I then have several minutes after I’ve pulled my blades from the oil to getting them into my press to flatten them.
Most carbon blade steels will harden satisfactorily in a fast speed quench oil. Exceptions are ‘W’ grade steels W1, W2 etc which are designed to be water quenched and plain carbon steels below 0.7% carbon (1055, En8 etc) but these are not really good blade steels anyway. I have quenched 6mm thick 1070 steel, which probably ranks as about the lowest hardenability of steels commonly used for blades, quite successfully in a fast oil.
The TTT diagram shown here for 1095 steel shows how the knee of the curve is very near the vertical axis and to fully harden the steel is virtually impossible. However, using fast speed quench oil will give a sufficiently fast quench to give a good working hardness on 1070 & higher plain carbon steels. Steels such as 5160 and En45 use other alloying elements such as chromium and silicon to move the knee to the right and so increase hardenability so good hardening will still be obtained even though they have a lower carbon content. However, the potential to straighten blades during the hardening process (as described above)is virtually non-existent (as shown by the red area).
I do NOT recommend that you agitate the blade in the quench. Your quench tank should be deep enough to lower the blade in vertically (my quench tank is a piece of 150mm dia pipe about 800mm deep). If you try to swirl the blade around whilst it’s cooling you stand a good chance of causing it to distort. It’s better to just be able to hold or hang it still while it cools. The blade section is thin relative to its mass and the heat will transfer to the oil quite fast enough providing you have the right oil.
If you’re treating several batches of blades then it will be necessary to agitate the oil between batches so the oil has an even temperature throughout prior to the next quench and don’t quench if your oil is over 80-900C (lower quench speed and fire risk). If you mass produce blades then cooling of your quench tank may be a necessity.
I hope this all makes sense and please do not hesitate to contact me if you need any further information.