Edge quenching

C

catcalls

Joined
May 7, 2012
Messages
250
I have been just quenching the edge of my blades recently, after thinking about how diferentially hardened blades are a lot more flexible(?!?!?!) So my question is; is an edge quench actually that much better than full-blade quenching? I have actaully not had any warpage since I began edge quenching. What are your guys' takes on this?
Jack

Thanks in advance
 
Depends on what steel you are using also a 5160 or 52100 steel blade will harden some what in the air, clay will help keep the spin hotter longer so it will give you a little more flexability. But edge heating and edge quenching gives the best results for smaller blades clay works best for swords or larger blades and edge quenching works best when you are going to be batoning a lot and do not want dents in the spin over time. Clay and edge heating are best for flexability.
 
Each process leaves you with a different microstructure. My preference is a homogenous hardening process(full quench) with tempering to suit the intended function(Full or differential, depending...). That is not to say that edge quenching or differential heat treating doesn't have a place in knifemaking.

Flex is a product of geometry within the elastic range... not the absence of hardness. If there is any "flex" beyond the elastic limit(which is the same for like geometries, regardless of hardness) it is a result of hardness(martensite) and the resistence against plastic deformation(strength) not the ability to deform without breaking(toughness).

I may have butchered that last bit of technical talk(Modulus of Elasticity)... lol. Page will probably come along and slap my peepee.

All I know, is that I would rather have a blade that requires 2 guys and a cheater bar to break, than one that can bend back and forth under my own power. I abuse my knives and don't mind that they get abused buy my customers. It is a compromise for me. Some would never expect that from a knife(by standard definition)... which IMO, makes edge quenching/differential hardening an even harder sell... from a practical standpoint.

There are good makers who use different methods for different reasons and as long as they aren't misrepresenting facts, I am fine with it.
 
Last edited:
Rick Marchand said:
Each process leaves you with a different microstructure. My preference is a homogenous hardening process(full quench) with tempering to suit the intended function(Full or differential, depending...). That is not to say that edge quenching or differential heat treating doesn't have a place in knifemaking.

Flex is a product of geometry within the elastic range... not the absence hardness. If there is any "flex" beyond the elastic limit(which is the same for like geometries, regardless of hardness) it is a result of hardness(martensite) and the resistence against plastic deformation(strength) not the ability to deform without breaking(toughness).

I may have butchered that last bit of technical talk(Modulus of Elasticity)... lol. Page will probuably come along and slap my peepee.

All I know, is that I would rather have a blade that requires 2 guys and a cheater bar to break, than one that can bend back and forth under my own power. I abuse my knives and don't mind that they get abused buy my customers. It is a compromise for me. Some would never expect that at all from a knife which makes edge quenching/differential hardening an even harder sell... from a practical standpoint.
Click to expand...

This is why i use 5160 and fully heat the blade and edge quench this gives around 55 hrc spin and 59- 60 hrc edge. Less flexible but slightly more the a fully hardened blade love this heat treat.
 
Rick Marchand said:
Each process leaves you with a different microstructure. My preference is a homogenous hardening process(full quench) with tempering to suit the intended function(Full or differential, depending...). That is not to say that edge quenching or differential heat treating doesn't have a place in knifemaking.

Flex is a product of geometry within the elastic range... not the absence hardness. If there is any "flex" beyond the elastic limit(which is the same for like geometries, regardless of hardness) it is a result of hardness(martensite) and the resistence against plastic deformation(strength) not the ability to deform without breaking(toughness).

I may have butchered that last bit of technical talk(Modulus of Elasticity)... lol. Page will probably come along and slap my peepee.

All I know, is that I would rather have a blade that requires 2 guys and a cheater bar to break, than one that can bend back and forth under my own power. I abuse my knives and don't mind that they get abused buy my customers. It is a compromise for me. Some would never expect that from a knife(by common definition)... which makes edge quenching/differential hardening an even harder sell... from a practical standpoint.

There are good makers who use different methods and as long as they aren't misrepresenting facts, I am fine with it.
Click to expand...


Wow man that us a mouthfull and the very thing I was pondering today after flexing a 52100 blade! Thanks
 
Ok, Ok... now I see! Mr. Marchand... that was some of the most complecated knife stuff I have ever seen!!!
Although I kinda understand what you're saying. Thanks to all who replied.
Jack
 
Ok, Ok... now I see! Mr. Marchand... that was some of the most complecated knife stuff I have ever seen!!!
Although I kinda understand what you're saying. Thanks to all who replied.
Jack
 
Rick Marchand said:
Each process leaves you with a different microstructure. My preference is a homogenous hardening process(full quench) with tempering to suit the intended function(Full or differential, depending...). That is not to say that edge quenching or differential heat treating doesn't have a place in knifemaking.

Flex is a product of geometry within the elastic range... not the absence of hardness. If there is any "flex" beyond the elastic limit(which is the same for like geometries, regardless of hardness) it is a result of hardness(martensite) and the resistence against plastic deformation(strength) not the ability to deform without breaking(toughness).

I may have butchered that last bit of technical talk(Modulus of Elasticity)... lol. Page will probably come along and slap my peepee.

All I know, is that I would rather have a blade that requires 2 guys and a cheater bar to break, than one that can bend back and forth under my own power. I abuse my knives and don't mind that they get abused buy my customers. It is a compromise for me. Some would never expect that from a knife(by standard definition)... which IMO, makes edge quenching/differential hardening an even harder sell... from a practical standpoint.

There are good makers who use different methods for different reasons and as long as they aren't misrepresenting facts, I am fine with it.
Click to expand...

you kind of got the whole flex thing right, so easy a caveman can figure it out :D

One of the most enlightning moments I have ever had was Tim Zowada and Kevin Cashen's demonstration at Ashokan of flexibility compared between a fully annealed and a fully hardened bar of O-1.

The setup:
2 bars cut from the same parent bar of O-1 precision ground spheroidized annealed tool steel
rigid mounting for bars with a strain gauge to measure bending force
lever for applying bending force to the bars of steel

The demonstration:
one bar was fully hardened but not tempered, the other left as is
first the annealed bar was subjected to bending force and the force required to deflect it was recorded as a graph of force against deflection. The graph climbed then fell off dramatically as the force exceeded elastic yield and the bar bent like a noodle

The fully hardened bar was put in the fixture and identical force applied. The force required to produce a given amount of deflection tracked identically to the annealed bar right up to the point where the annealed bar yielded, the hardened bar continued to climb steadily until it suddenly snapped.

The point of this demonstration: flexibility of steel is a function of geometry and cross section, nothing more. Hardening it changes where the yield and failure points intersect the curve, tempering introduces a yield point before failure (OK it is a bit more complicated than that but for this discussion let's ignore brittleness)

As either Tim or Kevin said (I forgot which, it was a couple years ago) a razorblade is an extremely thin strip of steel hardened to RC60. It can be bent double and will spring back to straight. If you have a very thick blade and you bend it, you are in effect stretching the outside radius leveraged against the inside radius, what causes the blade to fail is when the force on the outside radius exceeds the tensile strength or nonelastic yield of the steel, at which point the blade either breaks or bends. If you want to make a blade that will not break under a given load, make it thin enough to bend before the stress on the outside radius exceeds either limit, or if you want to make a crowbar, increase the cross section accordingly

-Page
 
Edge quenching introduces a whole extra layer of problems with no benefit (getting back to the OP of the thread). when the edge gets quenched while the spine doesn't you get a soft spine, (possibly pearlite) with a transition zone that is almost certainly pearlite, which has little tensile strength, the edge is brittle, it makes a fulcrum which helps tear apart the pearlite, The soft spine yields prematurely rather than supporting the edge. You are much better served by full hardening the blade and tempering it, while thinning the blade appropriately, if you need a further temper on the spine because of planned abuse like batoning the blade, draw a deeper temper on the spine

-Page
 
Many thanks to Page and Rick for taking the time to provide some serious clarity. This is an interesting but typically murky conversation around here...

I appreciate you guys.
AWL
 
I now understand more... however, would it be a better route to fully harden the blade, then use a torch to just heat the spine, or would this have the same effects as just edge quenching? When the spine is left in the state of pearlite(?) during edge quenching, would the spine of a clay hardened blade also be pearlite? Thank you to all who have responded.
Jack
 
catcalls said:
I now understand more... however, would it be a better route to fully harden the blade, then use a torch to just heat the spine, or would this have the same effects as just edge quenching? When the spine is left in the state of pearlite(?) during edge quenching, would the spine of a clay hardened blade also be pearlite? Thank you to all who have responded.
Jack
Click to expand...

From what I have read differential tempering will leave you with a harder spine than edge quench or a quench with clay. i.e. an oil quenched blade out of 1095 will leave you with martensite, retained austenite, and maybe some pearlite. If I remember correctly the reported hardness HRc of 60 - 64. A 400 degree F temper dropping the hardness to 58 - 60. Drawing the temper along the back to a green - brown - straw color 550 - 450 fahrenheit will lower the HRc from @ 5 to 1 points. So your spine could be @ HRc of 53 while your edge is 58 HRc. With the pearlite the spine will be much softer. I hope those with more experience will chip in.
 
catcalls said:
I now understand more... however, would it be a better route to fully harden the blade, then use a torch to just heat the spine, or would this have the same effects as just edge quenching?
Click to expand...
No it is not the same. I believe that a differential temper(torching the spine) is a better way to go than not hardening it in the first place. Martensite(result of a full harden) is stronger than pearlite and tempered martensite can be made very tough. To me, it just makes more sense from a metallurgical standpoint. But like I said, everything has it's place. If you require your blade to bend(as in the ABS performance test), you may find that edge quenching is the way to go.

catcalls said:
When the spine is left in the state of pearlite(?) during edge quenching, would the spine of a clay hardened blade also be pearlite?
Click to expand...
Stacy, Mete and Page could answer this better than I, as I have never taken to clay treatment enough to find out exactly what structures are formed. I would think it would be predominantly pearlite with some retained austenite and random matrensite because it is cooling from austenite at a fairly rapid rate(but not enough to harden.)

If I had to choose one of the three(torch, edge quench or clay), I would think that the clay treatment is the way to go, because IMO you could have better control of heat and quench rate than with a torch or edge quench.
 
Last edited:
A clay coated spine on a steel like 1095, W1, W2 etc will give you a mostly pearlite spine. What we see as a Hamon is where the steel is changing structures. The wisps and clouds area is where the mixed structures occur.

I had a really hard time coming to grips with the fact that you CAN NOT change a steels flexability by anything other than changing its geometry. All we can change is when and how it will fail when flexed.

If I was really after softer spines all the time I think I would make up a set of copper bars that clamped to the edge of the blade and set in a pan of water holding the blade with the spine exposed . Copper transfers heat very well. Then I would torch the spine with an oxy acetylene torch am mid tip monitoring it with my infrared gun or some heat crayons. Another thing is know your steel because many have a second hump in the temper where the steel gets harder again. Like A2 its hardness drops to around 850f then climbs from there to 950 and then starts to drop again.

I have done the edge quench (when I started) some clay coated (love hamons) and drawn back spine. I understand why some make extreme use knives like Rick likes to work on. But, for what the vast majority of people are actually doing with a knife picking the right steel, with the right initial heat treatment will accomplish everything needed. If I wanted a survival knife I would probably use something like 5160 at around 58 RC with a drawn back spine. In a survival situation I would be pretty careful with my only knife even if Rick made it.
 
The info from Page and Rick pretty much says it all.


Simple answer:
Martensite makes a better blade than pearlite
Martensite is stronger than pearlite
Martensite is harder than pearlite
Martensite resists permanent bending more than pearlite
Why would you want pearlite in your blade?


A more complete answer:
Flex is a product of geometry. The thinner, the more flex....it is that simple. Doesn't matter how you HT it, or which steel type you use.
Bending and staying bent ( exceeding the elastic limit) is a product of the steel structure and the hardness. Martensite will resist this more than pearlite.
Breaking is when the elastic limit is exceeded to the point of catastrophic failure. Unless the blade is very poorly heat treated, this should not be an issue in normal knife use.

Edge Quench vs Drawn Spine:
Edge quenching causes a very distinct boundary between two different structures - Pearlite in the upper part, and martensite in the edge. They bend differently. This can lead to either the knife taking a permanent set if bent hard, and/or the edge cracking.
In an all martensite blade, with the spine is drawn to a higher temper ( softer), the whole blade is one structure, and gives a more even reaction to bending. It will resist permanent setting just a little more. If over bent, the martensite blade may fail (break) sooner that the edge quench blade, but the edge quench blade would also be permanently bent at the same point ( with a possible cracked edge). But, remember that we are talking about stresses far beyond any normal use to get to this point.

Clay Coating:
When a blade is clay coated, you take control of roughly where the steel will be softer. The (shallow hardening steel) blade is fully austenitized, and then quenched in a fast quenchant. The exposed edge quickly hardens into martensite, but the slightly shielded spine misses the pearlite nose and goes into fine pearlite. The area between the two becomes a mesh of martensite plates and pearlite sheets ( old name was troosite). Looked at from the side under magnification, they seem to be a pile of needles, and are often improperly called that. This makes a very strong junction of the two structures. The martensite that forms in this region gets auto-tempered by the hotter pearlite steel that cools slower under the clay. The result is a somewhat controlled mixture of structures with areas of martensite and pearlite strategically placed. This makes the ashi and hamon lines we see on the surface, and makes structural bracing inside the blade.
Before anyone gets the impression that this makes a superior blade, it doesn't. A modern metallurgy steel blade with a well done HT to make 100% martensite will be a far better blade. The process of yaki-ire ( clay coated edge hardening) was developed to deal with more primitive metallurgy, equipment, and skills in a time when swords had real use. A bent sword was preferable to a broken sword. There were two ways to get that - 1) make the sword softer with a blunter edge; 2) Make the spine softer and the edge harder and better cutting......which one would you want in a battle?
The aesthetic aspects remain the main reason to continue the skill of attaining a hamon by clay coating the spine.....that and it is just cool to do.
 
As these processes and terms are a mouthful, I wasn't quite getting the whole metallurgy terms, however when it was said a little bit simpler I started to understand more and more. I now know a LOT more than I did on this subject, and thanks to all who took the time to share your knowledge and experiences.
Jack
 
I'm learning a lot from this thread, especially Stacy's recent summation. If I am to interpret correctly, the consensus is a thinner geometry, martensite edge, fine pearlite transition, from slow a drawback blade will yield better under lateral stress? If this is true, than a thinner blade (possible 3/16" or thinner) with a wider profile fully hardened and then drawn back....would be the best bet for the J.S. performance test?
 
You must log in or register to reply here.
Back
Top