View Full Version : In need of tempering help

04-20-2011, 07:47 PM
I am trying to build a boning knife. The blade is about 6.5 inches, CPM 154 and thickness of about .100. I HT'ed it in the "profiled only" state and got 62-63 Rc out of the plate quench. Now I am trying to temper to soften it more than I would any other type of blade but can't seem to soften it much. I typically temper at 400 deg, but that left the blade at about 60 Rc. I have been increasing the temper temperature gradually and just got through with a 600 deg temper. I am still getting 59-59.5.
How high can I go, or should I go to get about a 57-58 Rc? I want it to be as flexible as that thickness will allow, but also as tough as can be (I have broken one so far when playing with flex).
I would appreciate any help or would like to hear any experience with this type of build. Boning Knives are out of my norm.

Rudy Joly
04-20-2011, 08:39 PM
You should be able to go to 975-1000 degrees for 57-58 rc.
My info says a double temper at 950 degrees gives 59-60 rc.
My factory boning knives (Henkle & Sabatier) don't have a whole lot of flex, are you maybe refering to fillet blades ?


04-21-2011, 04:33 AM
I have seen info stating to temper in those high ranges, but that seemed to pertain to the "high critical temp" recipe. Every recipe I have found for the "lower critical temp" has shown tempering somewhere in the 400 deg range (which I am using).
I am finding it funny that a two hundred degree jump in temper temp hasn't really done anything to the hardness. Maybe I do have to go the the 900+ range. Is that o.k. to do with the low soak recipe??
The knife I was shown to make one similar had a decent amount of flex. It wasn't as much as a thin fillet knife, but more than what the first one I made was producing. I just don't want to put something in someone's hands that has a tendency to be "brittle".


Kevin R. Cashen
04-21-2011, 07:58 AM
“Flex” is one of the most commonly misunderstood topics in all of knifemaking. This is mostly because it is one of those things that your eyes are telling you one thing due to interpreting so many things at once, but when you separate out what is really happening the facts are an entirely different story. I am not saying that you may not understand the physics scherar but it is worth discussing for all those that may not know the facts.

When you pull a blade over in an arc there are several things involved- 1. how many degrees it deflects, 2. how much force it takes to deflect it that many degrees, 3. and what happens to the steel when it reaches that many degrees. You determine number 1, the thickness of the metal determines number 2, and heat treatment (hardness) can only affect number 3.

The yield point is where “flexing” ends and the metal bends. In flexing once you release the load the metal returns to true, in bending it doesn’t. It drives me crazy to see the number of makers who intentionally mislead the public by promoting their blades with outrageous “flex” tests when they are bending the darned things. The metals ultimate strength is determined by the point at which the blade breaks and this is always greater than the yield point. Think of the whole exercise of bending/flexing a blade as a line with breaking at the far end, and bending is a sliding point in the middle. All heat treatment does is slides the yield point (bend) up and down that line. The more you temper, the lower the yield point. The harder the steel, the higher the yield point as is gets closer to the ultimate strength. So a softer blade will bend at let’s say 30 degrees but not break. But a harder blade would not bend and continue to flex to 45 or 50 degrees and beyond but when it yields it will break.

In neither case can the heat treatment in any way effect the amount of force it takes to flex the blade to the yield point, that is an inherent property of the steel known as modulus of elasticity (or Young’s modulus) that is only affected by the amount or thickness of the metal. So if you are making springs and you want a weaker one, you grind it thinner, if you want a stiff spring you make it thicker, if you want a bent spring (not very useful) you make it softer. Before you start dismissing this, think about it – how do truck springs fail? Do they bend or do they break? How do you stiffen the suspension? You add more or thicker springs; you do not mess with the heat treatment at all.

Now the thickness of the metal can also affect the chances of yield or failure in a flex/bend by limiting the amount of tensile forces generated on the outside curve. Say your blade is 6” long and may fail at a 60 degree flex; this is because the thickness beyond the centerline transition from compressive to tensile action resisted enough to exceed the tensile strength of the steel for that length of blade. If you make the blade longer, or grind it thinner, you now will skew these numbers and the blade will be able to handle more degrees of flex without reaching the yield point or failing. This is why I always teach that the secret to beating the ABS performance test is as much in the grinding as it is in the heat treatment, and that somebody is going to get hurt someday with these terribly thick test blades.

Good springs don’t bend; they break, but only after extreme deformation. Good pry bars don’t do either and are thick to avoid even flexing. Good knives cut things and the physics that make the first two items work well can work against a good knife, if we don’t understand it all.

04-21-2011, 12:43 PM
This is just a quick response after glancing over yours during my lunch break. I'll have to re-read it a dozen more times when I have more time.
I guess I failed to show my "somewhat" understanding that steel thickness directly correlates with its ability to flex and has nothing to do with its hardness. I say this because you have responded to a post of mine before with similar aspects. I just have a hard time fully understanding that. For the most part it makes total sense, but I guess I have seen some simple instances that hinder my understanding.
First example: If you take a file and put it in the vise and flex it, it will break very quickly (I don't know the degrees). So that file probably left the factory in the mid 60's Rc? So, if that file was drawn progressively down in hardness, am I wrong to assume that it will still break at the same point? Once you get to a rockwell that is even lower (I don't know what that point would be), then the file will flex until it bends, then break? I haven't tested this progression, and you probably have, but it seems to me that as you keep dropping points, it will become less brittle.
I am not doubting your knowledge, I am just having a hard time seeing the whole picture. Bottom line, I am trying to finish this knife and not have it fail under normal use. I may have not read your response totally correct, but it sounds like I can make this knife as hard as a normal HT, and it will be as tough as one that is a few points lower. I have seen some of these knives listed (these are factory knives) with Rc's lower than most normal knives. Maybe I just assumed that a little lower Rc would help the problem.
I would appreciate it if you would continue with this. With your responses and testing on my part, this will definitely help my understanding of the concept.

Kevin R. Cashen
04-22-2011, 09:15 AM
A file is a very non standard thing to use to illustrate these concepts since it has a surface covered with defects that would make the outcome extremely unpredictable. Using a flat, smooth piece of steel at 63 HRC you can flex it to any degree under the yield point and the amount of force required to flex it will be exactly the same for any other identically shaped piece of steel regardless of the hardness. At 63 HRC the steel will have a yield point extremely close to its ultimate strength so there will be hardly any perceptible bending before the piece snaps. Take the same steel and draw it back to 55HRC and it will take the exact same amount of force to flex it under the yield point but the yield point will occur much sooner and the piece will bend appreciably before it thinks of breaking.

We did a test at Ashokan one year that illustrated all this quite well. Tim Zowada brought a gadget he made and two sections of the same bar of O-1, one 62.5 HRC and the other in the annealed state. The steel pieces were clamped at one end horizontally, side by side, with weights hanging off the free ends. Both deflected identically under 15 lbs. of force. This was all under the yield point in the flexing which is known as the "proportional range", it is called this because all of the deformation of the steel is directly proportional to the amount of force applied, more force more deflection and if you want to deflect it more you have to apply more force. At the yield point there is no more proportional force and so if the yield point is 20 lbs. the steel will continue to deform, or bend, with the same 20 lbs. of force, it is done.

At Ashokan the two pieces of steel were then mounted in Tim’s gadget that would measure any permanent yield and the exact force applied by a hydraulic ram. The annealed piece of steel had a proportional range up to around 50 lbs. at which point it took a set and no more force was needed to continue to bend. The 62.5 HRC steel took 450 lbs and then broke after a very slight bend.

These numbers could have been changed without using heat at all if we just would have ground the bars thinner, the amount of force to deflect would have dropped significantly and the degrees of flex would have increased as well since we would have moved farther away from the tensile strength of the material.

I would strongly suspect that the reason boning knives will have a lower Rockwell would be because the edge comes in contact with bone and cannot be chippy, rather than having anything to do with flex. Also Western cultures tend to go with really soft knives for kitchen cutlery and rely heavily on steels to continually realign the edge as butchers have traditionally done. In the East they tend to go with very hard and very fine edges but use them for very specialized tasks for fine slicing of soft mediums. Much of this is cutlery still catching up with alloying, most of our traditions in butchering came about when simple carbon steel was all there was and the only recourse you had was degrees of hardness to deal with the physics.

04-22-2011, 03:16 PM
Thank you for the response. Maybe my original question was a bit misleading, but I was trying to ask about tempering to increase toughness, not flexibility. I stated that I wanted it to be as flexible as the steel would allow (which I picked up a long while ago from your response to another post), but yet tough (maybe I am not using the right descriptive word for that).

I still don't think that I am fully understanding the concept. The reason I say that is because I interpret your response as flexibility is pretty much a measurement of the application of pressure. I understand that a certain dimension of steel will flex so much under X amount of pressure. But I guess what I am having trouble with is that even though two pieces of steel require X amount of pressure, regardless of their hardness, they will not flex to the same degree before failing. Will they?

So maybe the file example wasn't the best due to the stresses that could be found in them, but lets take the two pieces of O-1 you were referencing. If one was at +-63 and one was at +-57, you are saying that they will both flex the same amount before breaking? I can't say that I have actually tested this, but an example is when I snapped a blade right after the quench. I tried flexing it just a little and it broke without much flex. It probably took just as much pressure to flex it as another piece of that dimension, but it broke very quickly. Now after tempering that same blade, it would still take the same amount of pressure to flex it, but it would flex more before failing. Does my reasoning make sense, or am I seeing this all wrong? A blade right out of the quench is very brittle, a blade tempered is less brittle; a blade hardened/tempered to 60 Rc is more brittle than a blade hardened/tempered to 57 Rc, etc.

Once again I am not doubting your knowledge, I am just the type that is going to continue to ask questions until I get it.

Thank you for your time!

Kevin R. Cashen
04-22-2011, 06:39 PM
Introducing the term toughness does indeed change things and clarifies the topics. Toughness is most clearly measured under impact conditions and is very heavily affected by hardness or softness. Flexibility to me reflects the steels ability to flex, which would be easily discernable from bending in that the steel would return to it original shape when the load is removed, once again the deformation is entirely proportional to the force applied or removed. Flexing deformation is elastic, bending is plastic deformation. So flexibility to me would be the steels ability to elastically deform under any given load without either breaking or bending. I am sorry if it sounds like I am reiterating the same concepts but I want to help clarify where I am coming from.

A fully quenched and untempered blade will appear to snap quickly because there will be no discernable separation between the yield point and the failure point, thus it would give no warning and no pause, it would just snap. Tempering it back just a bit will allow imperceptible mechanisms in the steel to better cope with the strain but this gets off on another tangent regarding slip systems in elastic strain. This case is also similar to the file in that steels over .5% carbon should never be put into service without a temper and this is why, such a blade would not even need our outside force to eventually self destruct.

The question brittleness is also most evident under impact situations. A 60 HRC blade will resist plastic deformation much more than a 57 HRC blade but it will break with much less bend. The 57 HRC blade will take a slight set under less force. But put sudden loading such as impact on the two and the 60 HRC will snap under very few foot pounds but the 57 HRC blade will handle a few more. Large chopping blades benefit from impact toughness much more than small slicing blades, since the strength and abrasion resistance necessary for edge holding can be at odds with toughness.

Since most western users are quite accustomed to soft butchering knives that work with frequent dressing from a steel, I would go with your 57 HRC goal. It would be better to address the expected use that it is likely to see rather than throw anybody a curve ball. I am often re-educating many of my customers these days regarding small slicing knives. I wish to give them am incredibly sharp and long lasting fine edge but I need to first address some of the ideas about the kind of abuse any knife can see due to some of the deplorable marketing over the years. I really wish my friend Roman Landes’ book had an English printing since he has a portion of it devoted to how great knives could be if they could be made for their specific tasks without having to compensate for needless abuse.

04-22-2011, 07:09 PM
I know that I am not the smartest guy in the world, but I am learning everyday through experience, questions and research. I am just glad that I wasn't way off in left field and everyone else was at the bar celebrating after the game. But this still leaves me with needing help.

Back to CPM 154. I HT'ed it with a 1450 short soak, 1900 final soak and plate quenched, as I do all of my blades. It came out great at 62-63 Rc. I usually temper at 400 and get about 59-60, but as I stated, I want to go a little softer for this blade. I have gradually increased my temper to 600 and still haven't really dropped much in points. How high of temper can one go with the "lower temp" HT for that steel? It seems like somewhere in the 400's is normal for this recipe. I have read for the "higher temp" HT that it calls for a high temperature temp (900-1000). This is where I am needing help.

Thanks again.