With modern steel, is a differential quench ever superior?

I've seen a lot of people recommend a 1/3 of the lower edge be hardened. I don't really aspire to that myself, I prefer about 1/2 or more the width to be hardened, it gives me the toughness and rigidness I require. And depends on which steel I'm using. I like heating the edge and doing an edge quench as it gives me more control than other methods I've tried, but that is a personal preference.

To really answer your question, make three knives as close to the same as you can, fully harden one, edge quench another, fully harden and soft back draw the third and test to destruction and see which one performs better for you.
 
William, I have half a dozen 1095 blanks that I CNC machined to rough finished dimensions. I'm never going to finish them because the design is poor, but they would be great to test as you describe since they're identical.
 
kuraki said:
The fully hardened sword in motion will have the advantage over a fully hardened or differentially hardened sword in stasis.
Click to expand...

This is not accurate. Try this thought experiment:

You're in a train car and you have a sword that is stationary relative to you and the train car but moving down the road at 40 MPH (because it's in a moving train). You now swing a sword into it at 40 MPH, swung in the direction opposite of the direction of the train. The swinging sword is moving at 40 MPH relative to you, the train car and the "stationary" sword, but because it is being swung in the direction opposite to the direction of the train it is stationary to the ground outside the train (and therefor to an outside observer watching from outside). When you think about it this way, which sword is stationary and which sword is moving? It's an arbitrary distinction and it doesn't matter. Inertia and the resulting forces are the same. As far as which sword is "moving" it makes no difference in the outcome.
 
Kuraki, that would be an excellent test! I'd suggest, since you have 6, that you fully harden two, give one a soft back draw, fully heat one and edge quench, heat only the edge of another and quench, and play with different quench depths for the remaining blades. Then after tempering, cut, pry, baton, and beat the crap out of them. If you get to the point of bending to 90 deg. I'd suggest using a torque wrench if possible to tell how much force each method took to flex and bend. Only downside is that it'll tell you just about everything you need to know about THAT steel, to truly know on another steel you'd need to repeat.

Should also mention that design plays a roll as well, different shapes and grinds and profiles will resist bending and flexing more than others, or will spread the load more or less uniformly. If you want to know how well a knife maker makes a knife, ask them how they test there blades.
 
I don't mean to start anything but I'm surprised no one has pointed out that a katana is not designed or intended to block anything, in spite of what you see in movies.

Someone that knows how to use one will avoid blocking another weapon of any kind, and instead deflect blows away without absorbing the impact full on.

Sorry for the derail.
 
For some real experience -- Back in the old days when I was a fencer I looked at all the broken blades I came across. The foils were poor quality 9260 .The cross section was a fairly sharp cornered rectangle. The sharp corners did wonders ! The parries and especially the beats did wonders to create nice notches which started fatigue failures with further hits.
Rounded corners would have greatly reduced breakage .
But proper tempered martensite would be best as far as structure.
 
Just to be clear I don't believe that drawing back the spine is better, I just mean that if you do want to have a differential heat treat that it is better to do that way.

I'm only speaking technically... I love to do differential hardening and etch on 5160 and hamons on W2, 1095, etc. It looks cool and the knives are still extremely solid!
 
Nathan the Machinist said:
This is not accurate. Try this thought experiment:

You're in a train car and you have a sword that is stationary relative to you and the train car but moving down the road at 40 MPH (because it's in a moving train). You now swing a sword into it at 40 MPH, swung in the direction opposite of the direction of the train. The swinging sword is moving at 40 MPH relative to you, the train car and the "stationary" sword, but because it is being swung in the direction opposite to the direction of the train it is stationary to the ground outside the train (and therefor to an outside observer watching from outside). When you think about it this way, which sword is stationary and which sword is moving? It's an arbitrary distinction and it doesn't matter. Inertia and the resulting forces are the same. As far as which sword is "moving" it makes no difference in the outcome.
Click to expand...

You are correct of course. Relatively speaking! :D

Stasis, stationary, motion in relative perception, snow ball from a car...motion is relative.

In that test it appeared the "blocking" sword was being constrained in a way that prevents "all else being equal." It was not allowed to react. I do question the assertion I made though, because I initially thought the sword was being constrained at two points and hit in the center, which to me is a disadvantage. Now I rewatched and it appears to have been fully supported along the entire length of the blade, which makes me think it should have had the advantage.

Not a very scientific test by any means.
 
Will52100 said:
Kuraki, that would be an excellent test! I'd suggest, since you have 6, that you fully harden two, give one a soft back draw, fully heat one and edge quench, heat only the edge of another and quench, and play with different quench depths for the remaining blades. Then after tempering, cut, pry, baton, and beat the crap out of them. If you get to the point of bending to 90 deg. I'd suggest using a torque wrench if possible to tell how much force each method took to flex and bend. Only downside is that it'll tell you just about everything you need to know about THAT steel, to truly know on another steel you'd need to repeat.

Should also mention that design plays a roll as well, different shapes and grinds and profiles will resist bending and flexing more than others, or will spread the load more or less uniformly. If you want to know how well a knife maker makes a knife, ask them how they test there blades.
Click to expand...

I think a test in one steel would be roughly analog to another, but agree to "truly know" I would need to test the steel I wanted to know.

I'm trying to come up with some tests that I could repeat for each blade, as well as conduct further tests on each blade without the subsequent tests being invalidated by damage occurring in the first test. Other than isolating different sections of each blade I'm having trouble thinking of any.

I would almost like to laser cut a pile of coupons and use them as blade analogs. In my opinion whether they are ground in the shape of knives or not is irrelevant as long as they all have the same physical dimensions and are tested in the same manner. That would allow say, 4 coupons of each to go through 4 individual tests. And I wouldn't have to spend a bunch of machine or grinder time making knives I intend to destroy.

If anyone has a suggestion on how to test 6 individual blades so that multiple tests don't invalidate the results of each other I'm all ears.
 
duurza said:
I think under extreme stresses, both sections will act independently of each other. Not sure if the method they used to produce this katana and euro longsword were accurate but I think this video shows clearly what happens.

https://www.youtube.com/watch?v=w_3W1zg683A

Skip to 5:40.
Click to expand...

[snip-edit]

I think differential hardening could be advantageous in a very long blade that is subjected to impacts and shock waves, maybe not so much for a knife.
 
Last edited:
I think a test in one steel would be roughly analog to another, but agree to "truly know" I would need to test the steel I wanted to know.

I'm trying to come up with some tests that I could repeat for each blade, as well as conduct further tests on each blade without the subsequent tests being invalidated by damage occurring in the first test. Other than isolating different sections of each blade I'm having trouble thinking of any.

I would almost like to laser cut a pile of coupons and use them as blade analogs. In my opinion whether they are ground in the shape of knives or not is irrelevant as long as they all have the same physical dimensions and are tested in the same manner. That would allow say, 4 coupons of each to go through 4 individual tests. And I wouldn't have to spend a bunch of machine or grinder time making knives I intend to destroy.

If anyone has a suggestion on how to test 6 individual blades so that multiple tests don't invalidate the results of each other I'm all ears.
Click to expand...

I'll agree that they are analog to one another, with a caveat that some steels air harden even though they are oil quench. 52100, 5160, O-1, and L6 come to mind. We are not talking full hardness as quenched in oil, but unlike 1095 there spines will be harder than annealed if fully heated and edge quenched. I'd say that 1095 would be a good indicator of the simple 10xx steels, while 5160 or O-1 would be a good indicator of higher alloy steels.

I like the idea of the coupons, but in effect it would really only give you an idea of flex and difference in bending or breaking force required, a valuable thing to know for sure. A sharpened knife will tell you if it's ridged enough to withstand bending while using it to pry with since the thickness will be thinner and the geometry and distal taper react differently than a rectangular bar of steel the same thickness throughout.
 
You must log in or register to reply here.
Back
Top