I'm not sure what steels you are using (or whether it matters), but most of the literature I have on tempering the steels I use, (O-1, 1084, 440C, 154CM) calls for tempering at one temp for a minimum of an hour and a slightly higher temp for another hour (minimum). The temps change depending on the desired hardness and the soak time is for each inch of cross section, i.e. an hour per inch of cross section. So, if the cross section is less than one inch (typical for blades) then the minimum soak time of one hour at each temp is used resulting in a two-hour tempering cycle. I haven't seen tempering data that requires a longer cycle in a cross section less that one inch. Hope that answers your question.

While it seems intuitive that a shorter draw back time is sufficient for the thinner cross section typically found in blades and while this is true in general, I would caution anyone to err on the long side as far as time in the oven. There can be variables that we can not be aware of that will be addressed inherently in a longer draw.

Your question:

Do the end results justify the mean? By this I mean if the blade performance and RC test are suitable.

I would say Yes, but This question implies that some testing was used to determine that it lives up to some level of expectation. Testing is going to be your best way to dial in each individual combination of steel, cross section, time in heat, intensity/means of heat, etc. This testing will also be the way to get your best understanding of this procedure. The charts get us in the ball park so that newbies can make a knife that performs, but the testing and confirming of what we all read is what lets us understand.

|quoted:

While it seems intuitive that a shorter draw back time is sufficient for the thinner cross section typically found in blades and while this is true in general, I would caution anyone to err on the long side as far as time in the oven. There can be variables that we can not be aware of that will be addressed inherently in a longer draw.

Your question:

Do the end results justify the mean? By this I mean if the blade performance and RC test are suitable.

I would say Yes, but This question implies that some testing was used to determine that it lives up to some level of expectation. Testing is going to be your best way to dial in each individual combination of steel, cross section, time in heat, intensity/means of heat, etc. This testing will also be the way to get your best understanding of this procedure. The charts get us in the ball park so that newbies can make a knife that performs, but the testing and confirming of what we all read is what lets us understand.

Good advice Lin. It is always difficult for younger smiths to test a blade to destruction (I had a lot of trouble bringing myself to break blades)because we have too much time and money (relative terms I know) invested and we just want to finish that knife and try to sell it. However, you never really know how good your HT cycles are until you break a couple of blades and check the steel. I found that shorter tempering cycles than one hour resulted in different grain structures between the thinner sections and the thicker ones. Surprisingly, the thicker sections had smoother grain than the thin ones. The blade areas looked lumpy and coarse while the tang was more homogenous. I have recorded what produces a fine, smooth grain for the whole knife and use that for my O-1 blades.

All,

Tempering is a time and temperature function. That means you must do it at the right temperature for a sufficient amount of time. Why?

There are several reasons to temper: reduce brittleness and transform retained austenite are two main reasons. Which translates to I do not want it to break when I drop or use it and I want the edge to hold its sharpness longer. We all know this functionality related to tempering. If it is too hard the edge chips and if it is too soft it yields to the force of cutting into something.

At the molecular level "high carbon steels (0.3-1.5% C) are not stable at room temperature because interstitial carbon atoms can diffuse in the tetragonal martensite lattice at this temperature" (Key to Metals, 2010). So we temper to stabilize or change the slightly off tetragonal structure of the steel matrix to a more stable form. I like to think of this as getting this frozen (quenched), somewhat unstable structure to relax a bit so it can handle more stress when I use it.

Tempering also causes retained austenite to transform into primarily bainite, ferrite, and cementite. Hopefully, most of the austenite transforms into bainite (lower bainite). Regardless, austenite is undesireable at normal temperatures where a knife is used. Did I mention that we have untempered bainite? So, we put it back in the oven again to complete the tempering process and temper any as yet not affected martensite and to temper the bainite. We lose some carbon out of the martensite in the tempering process as some carbon migrates out of the structure as we add energy with heat. While we lose some hardness we gain toughness.

When we draw the back of a blade, we are tempering that portion of the blade. We do this at a higher temperature with our torch and by doing it twice or more we allow enough time and heat to temper that section of the blade so it can withstand more structural force. We can bend it further without permanent damage to its structure. Drawing the back of the blade is tempering process.

Dan P.

Sorry, I forgot to follow up on the initial question of whether less time is needed for a thin cross section of steel rather than a thick piece. Tempering involves the changes described in the my previous post and those mostly occur at the tempering temperature for the specific steel. W2 would be tempered at 450 F as an example. While some changes occur as the blade heats up past a couple of hundred degrees, most of it occurs at the selected temperature. Theoretically, it takes some time at that temperature to acquire the desired results. So the relative thickness of a 1/8 inch knife blade versus one that is 1/4 inch should be about the same time. Another way of looking at this is if I put both the 1/8 inch blade and the 1/4 inch blade in the same oven for the same amount of time at the same temperature they should not vary in hardness or for the most part internal structure. This assumes both steels have the same martensite structure to start with.

Dan