How to Test your Heat Treat

Casey Brown

Well-Known Member
Outside of a Rockwell Hardness test, what are some good ways to test to see if you got a good heat treat? I know about testing with a file to see if it skates or bites as a simple test. What are some other methods that you use to test your heat treat? I am currently heat treating utilizing my propane forge, and think that I am zeroing in on a good heat treat, but want to verify the quality of my heat treat before I even think about selling a blade to a customer.
 
Personally, I use the brass rod test on EVERY blade that goes out the door. It's not an indicator of any specific/exact hardness, but rather a test of the correct working hardness for the given steel, with the given edge geometry you've installed.

Don't fall into the mindset of "harder is better"! It's not! What you should shoot for is the correct hardness for the steel you're using, with the geometry you've ground on it.

Personally, I feel way too many knifemakers who only seek high hardness, forget that their clients/customer will eventually have to re-sharpen any blade that is used.... and if a client/customer struggles to, or flat out can't sharpen it because the hardness level is too high......it give the person who made the knife a "black eye".

I was caught up in the "high hardness" thing for a while during my early days, Rc testing nearly everything. After a while it became so predictable that I sold off the Rockwell tester...... it was just sitting in the shop gather dust and rotting. :) What hardness level do I use? For the steels I use, and the geometry I apply, 57-59 is what I choose. I don't demand those exact numbers on everything....various designs/uses dictate.
 
Personally, I use the brass rod test on EVERY blade that goes out the door. It's not an indicator of any specific/exact hardness, but rather a test of the correct working hardness for the given steel, with the given edge geometry you've installed.

Don't fall into the mindset of "harder is better"! It's not! What you should shoot for is the correct hardness for the steel you're using, with the geometry you've ground on it.

Personally, I feel way too many knifemakers who only seek high hardness, forget that their clients/customer will eventually have to re-sharpen any blade that is used.... and if a client/customer struggles to, or flat out can't sharpen it because the hardness level is too high......it give the person who made the knife a "black eye".

I was caught up in the "high hardness" thing for a while during my early days, Rc testing nearly everything. After a while it became so predictable that I sold off the Rockwell tester...... it was just sitting in the shop gather dust and rotting. :) What hardness level do I use? For the steels I use, and the geometry I apply, 57-59 is what I choose. I don't demand those exact numbers on everything....various designs/uses dictate.

Ed do you simply run the edge along the brass rod and see if it returns to straight. I have a small neck knife Im working on and it is aebl super thin and i can run my nail along the unsharpened edge and it kind of wants to defore a little.
 
The brass rod test, while valid in some senses, can be a little misleading too in others. Specifically, if you have to push REALLY hard to get your edge to flex or can't see it flex at all no matter how hard you push, then your edge geometry is likely too thick and the brass rod test won't mean much.

Personally I feel the brass rod test is more valid to a maker that KNOWS FOR CERTAIN their edge geometry is already dialed in.

For a new guy, I'd use a knife of already known decent quality to compare my knife's slicing and cutting capabilities on every day materials.
 
So, that is a good point. I'm definitely still playing with edge geometry. Trying to find out how thin to grind before it's ready for an edge is always kind of a guessing game for me, especially depending on the type of blade I am trying to make. I appreciate all of the feedback though. I haven't actually started trying to sell any of my knives yet, as I want to make sure that everything is "right" before I get a customer. I'm literally on my 21st knife that I have made, but only the second that I have actually done the heat treat on. I was actually thinking about selling this one. I think I got the heat treat figured out, but want to make sure before I let anyone spend money on something that represents me.
 
John makes a valid point about the edge geometry..... sometime I don't think outside my own "box"...... I strive to create blades with the finest edge geometry possible, and still have the edges hold up well under heavy use. That in itself dictates that I tailor the heat treatment of the given steel to the geometry I apply/grind.

Personally, I would say that if you can't use hand pressure to "flex" and edge over a brass rod, then you need to rethink how "heavy" your making the blade's edge geometry. Just to qualify that statement, I am speaking to those particular steels that are generally used when forging. You can't apply the same logic/standards to stainless steels....they just don't work that way.
 
i cut with it, i sharpen to a ridiculously low edge angle 7-10dps and then do kitchen work or whatever with a test knife, then I see how it degrades... for everyday / outdoors knives i take a heat treat sample grind it and put a very fine edge and whittle / cut hardwoods and see how the edge lasts... i also see how it behaves under sharpening stones to see if the burr is easily avoided when sharpening and the edge is easily formed at very low angles... then i know the steel is fine, then you can adjust the edge for the application...
 
The brass rod test, while valid in some senses, can be a little misleading too in others. Specifically, if you have to push REALLY hard to get your edge to flex or can't see it flex at all no matter how hard you push, then your edge geometry is likely too thick and the brass rod test won't mean much.

Personally I feel the brass rod test is more valid to a maker that KNOWS FOR CERTAIN their edge geometry is already dialed in.

For a new guy, I'd use a knife of already known decent quality to compare my knife's slicing and cutting capabilities on every day materials.

I am glad to hear you mention this John. The brass rod test is very similar to the ABS bend test in that is it based upon flexing and bending, so is rather vague on what it is really measuring, and thus very open to misinterpretation. If you did the ABS test just enough to flex the blade, but didn’t measure the degrees the handle moved, and the blade neither bent nor broke, how much about the heat treatment would it tell you? How much about the blade geometry and overall thickness would it tell you? This is what must be sorted out when flexing any part of the blade. Truth be told, unless you damage the blade (i.e. bent or cracked) all you have measured is the amount of stiffness due to the material thickness. The heat treatment only determines how the flexing action ends and, to some extent, when. But the “when” is still heavily influenced by the thickness, regardless of the heat treatment, if the steel is thin enough not to generate the required resistance to load to exceed the yield point.

E.g. take a set of feeler gauges, one hard, one dead soft. As you get thinner in the stack there will be less chance of bending or breaking, regardless of the heat treatment, until you can go all the way around in a loop at .001”
 
If you did the ABS test just enough to flex the blade, but didn’t measure the degrees the handle moved, and the blade neither bent nor broke, how much about the heat treatment would it tell you? How much about the blade geometry and overall thickness would it tell you? This is what must be sorted out when flexing any part of the blade. Truth be told, unless you damage the blade (i.e. bent or cracked) all you have measured is the amount of stiffness due to the material thickness. The heat treatment only determines how the flexing action ends and, to some extent, when. But the “when” is still heavily influenced by the thickness, regardless of the heat treatment, if the steel is thin enough not to generate the required resistance to load to exceed the yield point.

E.g. take a set of feeler gauges, one hard, one dead soft. As you get thinner in the stack there will be less chance of bending or breaking, regardless of the heat treatment, until you can go all the way around in a loop at .001”

This should make perfect sense - especially feeler gauges as an example - but I'm still in the fog here.

Kevin hit on this on my edge quench question: "The heat treatment only determines how the flexing action ends and, to some extent, when. But the “when” is still heavily influenced by the thickness, regardless of the heat treatment..."

What am I missing?

I got " bending" and "thickness" but how heat treat changes it - or not at all....huh? Seems to me if its soft it wont snap... and the harder it is the more likely steel would snap. But maybe I'm wrong as I've seen structural steel from buildings that were broke b/c of earthquakes probably because they reached some yielding point - not only bending, but also twisting.

Sorry I missed this at Ashokan. Didn't have the time or money, but if I did it would have been worth it. I'll dig around on the site some more.
This has my wheels turning now.
 
This should make perfect sense - especially feeler gauges as an example - but I'm still in the fog here.

Kevin hit on this on my edge quench question: "The heat treatment only determines how the flexing action ends and, to some extent, when. But the “when” is still heavily influenced by the thickness, regardless of the heat treatment..."

What am I missing?

I got " bending" and "thickness" but how heat treat changes it - or not at all....huh? Seems to me if its soft it wont snap... and the harder it is the more likely steel would snap. But maybe I'm wrong as I've seen structural steel from buildings that were broke b/c of earthquakes probably because they reached some yielding point - not only bending, but also twisting.

Sorry I missed this at Ashokan. Didn't have the time or money, but if I did it would have been worth it. I'll dig around on the site some more.
This has my wheels turning now.

Mr. Mind, or Sound if that is all right, as I am good with Kevin as well,

A good starting place in getting a handle on these concepts that I most often recommend is to research "Young's Modulus of Elasticity". The lecture and demonstration that Tim Zowada and I presented at Ashokan this last weekend was quite successful in helping those who attended to understand how misunderstood stiffness, resilience and flexibility are among knifemakers in light of the prevalence of tests such as the brass rod and softening spines in hopes to increase strength or "springiness" And edge "flex" without any permanent deformation actually only tells you how thick or thin you have ground it. Heat treatment only decreases or increases the range at which the steel will show elastic strain (flexing and returning to true); the harder the steel, the greater this range. "Soft" steel replaces this extended elastic range with more ductile behavior with plastic deformation. At the end of both conditions on the stress-strain curve is actual failure where the steel does come apart, just one is sudden after much more applied load, while the other is accompanied with larger amounts of plastic deformation.
 
Heat treatment only decreases or increases the range at which the steel will show elastic strain (flexing and returning to true); the harder the steel, the greater this range.

Gotcha - I misunderstood what scott and, chris were talking about. My misunderstanding was HT had no effect on bending the steel - contrary to the whole principle of heat treating.

I have little clip point I made - cutting away everything I didn't want out of a circ saw blade - just for the experience and experimentation. The circ steel clip point blade was very thin at the point. It would stick into plywood when I threw it and then bent and stayed bent when I hit a knot. obviously the steel (I tested it and it was hardenable) was tempered for it's purpose as a saw blade, soft enough so it wouldn't snap. I learned that. About this time the conversation between scott and chris came up. I wondered why wouldn't a HT cause that tip to come back to straight. But I see now, I'm right, it would.

I also have two blades HT'd by me which are different thicknesses at the edge. One izza razor knife, the other I hope is a little bone breaker. This threads going to help me understand a little of what to expect. Like Chris I'm realizing how edge geometry as comes into play.

I didn't expect you to really answer Kevin, I was just quoting your content and talking into cyberworld. I thought you hadn't even made it back yet. Thanks for the second answer which contained some of what you had already said. Just ignore me if I annoy you some simplistic question I throw out there. My first name is Luke, feel free to refer to me as that if you'd like.
 
Hello Luke,

I guess what most knifemakers should remember is that if they want avoid a bent or broken knife from lateral loading, they should make it thicker. I personally recoil from this concept because I have trained my entire life to make the best cutting tool I can and have never needed anything more than a $5 crowbar for lateral loading jobs. I like strength in my blade materials and differentially hardening a blade is a compromise of that strength, for ductility. A ductile blade will bend so much sooner than a properly fully heat treated would break that there really isn't much point, aside from aesthetic effects with hamon. Ancient steels were so different from modern alloys that the compromise was about the only alternative, but today we are blessed with steels that allow us to have our cake and eat it too.

Impact toughness is another story however. Tempering a blade back further if it is going to see heavy chopping is a sound practice since sudden loading behaves entirely differently from gradual loading.
 
I am still wrapping my brain around this concept too. When I tell people about riding a motorcycle I tell them the tires only have 100% traction. If you exceed 100% traction your tires will slide. So, if you are at full lean angle in a right hand turn using 99% of YOUR available traction to hold the motorcycle in that turn and you suddenly discover the turn had a decreasing radius and you try to use an additional 5% traction for braking you WILL lose traction. The point is there are many factors which effect where the 100% traction mark resides. Rider weight, type/hardness of tires, tire wear, road conditions, etc. The overall failure point of a knife should be able to be viewed the same way. The knife can only take a 100% load if you exceed 100% the knife will fail (failure can be viewed as breaking or deformation-stay with me here). The caveat is that there are MANY factors which effect what equates to a 100% load for any given knife. There are many ways to manipulate that 100% mark and many of them differ when you switch to another type of steel. Heat treat, geometry, thickness, taper are among these things. So in my mind the goal here in a VERY SIMPLE statement would be, for any design made from a specific steel, you need to find out what combination of thickness, steel, taper, HT, grind, edge geometry etc, will give you the results you deem to be acceptable for your knife. For some of this we can rely on what others have found to be acceptable to them but in the end the only way to be sure is to create-test-evaluate-change-repeat until we are happy with our creation. Am I even in the ball park here?
 
So not to high jack the thread. I have a 24" wakizashi .220 thick stock from 80crv2. Right now it is best treated and tested to a 58 hrc. I expect to use this for hard impacts chopping. Kevin do you think it would be wise to draw some of the harness out of the spine or leave it fully hardened at 58?
 
The knife can only take a 100% load if you exceed 100% the knife will fail (failure can be viewed as breaking or deformation-stay with me here). The caveat is that there are MANY factors which effect what equates to a 100% load for any given knife. There are many ways to manipulate that 100% mark and many of them differ when you switch to another type of steel. Heat treat, geometry, thickness, taper are among these things. So in my mind the goal here in a VERY SIMPLE statement would be, for any design made from a specific steel, you need to find out what combination of thickness, steel, taper, HT, grind, edge geometry etc, will give you the results you deem to be acceptable for your knife. For some of this we can rely on what others have found to be acceptable to them but in the end the only way to be sure is to create-test-evaluate-change-repeat until we are happy with our creation. Am I even in the ball park here?
I think so. being an old sailor, I like to add KIS(keep it simple). The more steps you add to HT, the more chances to make a mistake. The more you deviate from "Industrial standard HT", the more chances to make a mistake.
As Kevin said above, to avoid a bent or broken blade, don't use it for a pry bar. I cringe when I see videos of folks bending blades to more than 45* and because the knife did not break, claim their secret HT is the reason. the fact that the blade stayed bent and is ruined is never said. the blade bent it did not break so therefore my HT is best.
the best video I have seen about this, there are 3 pieces of ground flat stock, one soft, one as quenched, one tempered to Rc55. weights are place on the soft piece till it bends but does not spring back. the same amount of weight causes the as quenched to break and the tempered to bend but not spring back
 
I like to add KIS(keep it simple).

Yes, sir I feel the same> I test my blades but not in overly complicated ways. I try to keep them in the realm of what the knife MAY encounter when its in proper use. Actually I try to go one whole step above what the knife SHOULD see based on what it is designed to do. A razor sharp kitchen knife designed to thinly slice fish and veggies may get to cut a lot of cardboard to test its edge retention but I would never chop a 2 x 4 with it or bend it 45 degrees (unless I wanted to see what it took to break it). On the other hand a hard use field knife may get to chop through a couple of 2x 4's and then push cut paper. I do not chop car doors or nails to prove my knives do what they were designed to do. No problem if anyone tests that way its your choice. In my opinion a knife is a cutting tool designed to cut stuff. If I need to chop a tree down that is what my field axe is for, if I need to pry open a car door that is what a pry-bar was designed to do. I would never test a pair of scissors by attempting to cut steel with them because that is not their intended use. I think over time with the publicity given to "forged in fire" type testing we are going to harm ourselves by giving people a wrong sense of what a knife SHOULD be able to do.
 
Maybe when we start thinking of using knives beyond what they're intended for that's just pride. It's like saying, "Look at everything else my knife could do!" I could see myself trying to show off like that at camp or around the yard. Good to catch that now. I don't have a TV but relating to the few forged in fire shows I've seen, I could see toughness as essential for swords - you'd want it to hold up the abuse of battle (all day) and still slice.

For my uses it'd be cardboard, processing fish and meat. Bones seem to dull the edge on a lot of commercial knives I've owned before I got into looking to own/make a good blade. I'd like my blades to handle being scrapped across a bone again and again and still cut easily into the next animal. I could also see using to make kindling and cutting fresh willows. Mechanics wire in a pinch maybe. But then that's what multi-tools are for.

Not sure if this is helpful but I'll report on my testing... from new maker to new maker. Of these two I have I did some testing/cutting - everyday work with them. The thin one dulled on some spots cutting kindling. I could see a few shiny spot on the edge, which before was shaving sharp. But it still had enough edge to effortlessly slice cardboard. I think blade geometry played a major role in that, though. These blades also took permanent deformation from cutting mechanics wire. I might have drawn them back too far. Temps might have been inaccurate, though, I used a toaster oven with no firebrick or thermometer. The thicker one I had didn't like to cut cardboard even though it'd shave hair. But I liked its performance when I split a salmon backbone and cut off the jaws - only a total of six cuts. It held up to being it with a rock on the handle near the spine and went through salmon jaws like my hatchet.

When I get to my home shop I have known steel, and crude but controlled heat treating environment. I think that'll also come into play on the performance I get out of future knives.
 
Bones seem to dull the edge on a lot of commercial knives I've owned before I got into looking to own/make a good

Now in my opinion that would be a good test. Purposely SLICING into bone many times will tell you a lot about how your knife will function if it's intended to butcher. Just remember, this stuff is only my opinion, I could be wrong and surely some will disagree with me and that is OK.
 
Back
Top