Edge Quenching

One

Banned
I’ve wondered whether or not there might be more to the edge quench than what’s been stated and if traditionally it might actually have been more about edge quality and geometry than the whole 90 degree bend thing.

Heat treating a triangular or wedge shaped cross section is much different than heat treating a rectangular cross section, much more complicated and problematic because of the “temperature gradient“. So much so, that it is often actually considered a design flaw which should be avoided or at the very least dealt with. The problem is that there are different heating and cooling rates, differences in expansion and contraction etc., between the edge and the spine. In essence the whole edge is a stress riser.

Perhaps the edge quench is one way of dealing with this problem. Maybe it allows for higher quench speeds without as much possibility of cracking and minimal stock reduction post heat treat.

It talks about design influences on heat treating in this article.

http://books.google.com/books?id=SX... part geometry on heat treating steel&f=false

Any thoughts?
 
Thanks, Tai. Now that's one more book on my want list:20:. Good thing that it's cheap used.

Doug
 
Last edited:
You bet Doug.

What I'm thinking is that, there really isn't a good reason that a blade (meant for cutting) should need to bend 90 degrees or that anyone would want a blade that bends easily. Next to a blade snapping, bending would be the next worse. So, maybe there is a better reason for edge quenching, traditionally and otherwise. The edge is the primary working section of a blade, so it makes sense to give that part priority. Maybe the bending part was just a negative sort of trade off, in order to optimize the edge. However, this could have been compensated for at least in part by beefing up the geometry.

I know all the history books say the reason for differential hardening was to make the blade tougher and less likely to snap, but maybe that's not correct,... just a way for them to try and make sense out of it... or some type of confusion within traditional smithing camps that led to it.
 
Last edited:
For knives at least, could it be that it was just handier to use a shallow container sitting right next to the forge/oven? That's part of the reason I used it.

As for the temperature gradients from thicker to thinner sections during quenching, the gradients would be much greater/worse with half a blade in oil or water and the other half out.
 
As for the temperature gradients from thicker to thinner sections during quenching, the gradients would be much greater/worse with half a blade in oil or water and the other half out.

True, but maybe it would minimize the rapid split second shock and stress compared to quenching the whole blade. The back of the blade would act as a heat reservoir and maybe offer some auto tempering effect to minimize stress.

Quenching a smaller section or mass might help minimize stress to the edge during the quench.
 
Last edited:
I'm of the philosophy that if you need a knife, then get a knife, and if you need a pry bar, then get a pry bar. If you do a full hardening of the blade and then temper it, you will get a stronger blade than if you left the spine pearletic or bainetic but it would be less tough. Having a tougher blade might be an advantage with heavy chopping, though if you need an ax, you'd be better served with an ax. It just comes down to the trade-offs that blade design and production abound with.

Doug
 
Good points Doug,... so you have to wonder how much of the heat treating information we read on the internet is really geared more towards different geometries and different purposes than,... "knife specific" ?...

Heat treating knives is a specialization.
 
Last edited:
It would take a little investigation to see whether the extra heat in the spine would autotemper or just slow cooling down. If the latter, which I think is likely, it could have an effect on the purpose of the quench, ie. hardening. A great deal of the stress of quenching steel comes from the stress of the martensitic transformation. Precipitation hardened aluminum alloys are just quenched in water, in large part because there is no phase transformation from the quench itself. If the heated spine slowed down cooling enough to alter this transformation, then the stress would be reduced. However, if a knife is intended as a cutting tool, this would be an unacceptable trade off most likely.

The other source of stress in quenching is the thermal gradients discussed in the opening post. These will always be greater if part of the blade is cooling in a different medium than the rest, assuming the mediums don't by some chance have the same cooling ability. That split second of high gradient gets turned into several seconds in if the blade is edge quenched and the spine cools in air. If a soft spine is not desired, I think quenching spine first would minimize quenching stresses, giving the spine, with it's greater heat capacity, a head start, and evening things out as the thinner edge catches up. There is no way to eliminate these thermal gradients and even air hardened parts can crack.

Basically, it boils down to this IMO. I've seen the video of the katana quenched in a fish tank. Without a darn good reason, I'd want no part of all the ups and downs the blade does in that video. If the 5 or 6 inch long blade of a knife moves 1/4 to 3/8 of an inch or more, as reported by some who edge oil quench low alloy steels, and still doesn't crack, I think that points more to the ability of steel to tolerate edge quenching than it does to possible benefits of edge quenching, at least with respect to making the quenching process more tolerable.
 
My jaw dropped when I saw that video. It's surprising that anything survives what happened with that blade. As to whether or not there is autotempering during an edge quench would seem to depend on how it's done. I could see it if the blade is rocked back and forth in the quenchant until the spine looses it color and then removed. Less so if the if the entire blade is dropped into the quenchant at that point.

Doug
 
There’s actually a bigger issue at hand than the validity or invalidity of edge quenching…

The bigger issue is,… how might we be able to deal with the inherent basically triangular or diamond shaped cross sectional geometries of knife blades, thermal gradients etc.? I think there may be a number of ways.

Any ideas?…
 
I use the edge quench heat treatment for most of my large knives. I feel that the back never reaches critical temperature after forging and normalizing. The spine will get "hot" but below 1475, at least when I do it. I have always thought that I'm hardening the edge and normalizing the spine at the same time. From destruction testing I find the steel has a very fine grain and will take a lot of punishment especially impact.

Edge quenching produces a smooth hamon like line or zone that I like the look of, pure personal preference. I personally have never had a knife move when quenching in oil up to 10" blades.

Since this is such a controversial topic I will add the disclaimers
Edge quenching is not the only way to heat treat a knife just another way. There may be better ways or not I don't know, however through my personal testing I can make a knife that works perfectly fine using this method.
 
There are many ways to edge quench. My comments about thermal gradients are mostly directed at heatng all or a portion of the blade, then quenching a narrower section. As a heat treating method, it is the only way to accomplish some goals. However, in the context of this thread, it cannot minimize thermal gradients and the associated stresses. Not everyone reports movement when using edge quenching, just some.

Any technical information from the internet is questionable.
 
I think there are a number of variables to consider,... not the least of which would be quenching mediums (the associated alloys) and quench speeds. The book seems to suggest that. Modifying the geometry pre-quench is also mentioned.

How do we optimize the edge through heat treating etc., to give the edge priority?... In essence, shouldn't it all be "edge quenching” with knives?
 
Try thinking of it like this:

In a full quench scenario on a triangular cross section with a temperature gradient,… if the spine of the blade has been optimized, then the chances are that the edge has been over stressed.

Heat treating a knife blade for the spine would be similar in some ways to heat treating a ball bearing for the central core.
 
Last edited:
I think most heat treatments are chosen for the edge. When we talk about hardness, carbide size, grain size, edge stability, etc., those are all chosen for the edge. They don't hurt the spine, especially in a pure cutting knife, but they are chosen to give the edge the desired properties. Some knives have more overall concerns, so the edge may not be the first priority, but those are not the norm, not from what I've been reading on the various knife forums for the past few years anyway. Edge quenching can give what some smith's consider the optimal hardening for both spine and edge, but not every smith agrees what those should be.

As for methods to minimize thermal gradients and stress on the edge, there are several.

Marquenching
Spine first quenching
Quenchant choice and corresponding steel choice (water, oil, air, salt, etc)
Quenching after tempering
Grinding after hardening (though this presents it's own problems)
Leaving profile/geometry oversize before hardening
Interupted quenching
Vaccum/protective atmosphere
Foil wrap
Anti-scale compounds
Reducing forge atmosphere
Water cooled power sharpening or hand sharpening

All these things are intended to protect the blade, and primarily the thin section of the edge, from various effects that will harm it, such as decarb, overheating, uneven heating, uneven cooling, scale formation, etc. They also protect the rest of the blade too, but the primary concern is the edge.
 
Very good me2!... and there may be a few others.

So, it's not so important which method is "better" or "worse", but rather how the steel type or selection and it's quality, geometry, forming method and heat treating etc., all work together to make a superior blade.

The spine of a blade and the core of a ball bearing both serve their purposes of support etc., though more secondarily than primarily.

… Lots of possibilities and ways to solve geometric problems and concerns.
 
I must say, I don't really think that uneven cooling is that much of an issue for knives, especially with all the plate quenching, air hardening, and oil quenching going on nowadays. The greatest common spine thickness is about 1/4". Based on work by Verhoeven, we know that the cooling speed in the center of the thickest section is essentially the same as that at the surface and at the edge. Yes, the spine cools slower because of slightly greater mass, but I don't think the difference is that great. Knives are still a pretty basic shape compared to some of the stuff in the book exerpt above. It's not as though the cooling speed of the spine is dependent on the thermal conductivity of the steel itself, as is the case in thick dies (3-4 inches +). Yes there are cracks and warping and all manner of issues. There is also overheating, quench cracking, various flaws, and other causes for those issues. My worst warped and cracked blades came from edge quenching, but they could have warped from any number of causes based on my equipment and inexperience.
 
How much of an issue the triangular, wedged or diamond shaped geometries and their corresponding thermal gradients are with knives depends on a number of possible variables. The important thing is to understand the principals and how they might apply to knives.

Quenching blade geometries is different, more complicated and problematic than quenching, let's say, flat rectangular bars, due to thermal or temperature gradients. It is something than needs to be considered and dealt with one way or another… It is potentially a crack prone situation and just because a blade survives the quench doesn’t necessarily mean that it hasn’t been damaged or weakened by high levels of stress along the edge during the quench cycle.
 
Last edited:
Back
Top