Hardening II- Quenching

Should this thread be indexed?


  • Total voters
    44

Kevin R. Cashen

Super Moderator
Hardening II

Quenching


With the carbon in solution (austenite) within the blade, the idea behind hardening the steel is to somehow trap that carbon there to create a super-saturated solution at room temperature with the maximum wheel chock effect on the iron matrix. If the steel is cooled slowly the carbon will simply come out of solution to form concentrations of carbide within the iron, in its most common form this will be the lamellae of pearlite. To avoid this, the steel must be cooled at a rate faster than which the carbon can move and this is why we quench the steel.

At one time, when steel was just a simple alloy of iron and carbon, water was all that was needed to meet this task. But as demands on performance, along with the complexity of parts treated, increased we developed other quench methods and alloys to help facilitate them. With alloys that more readily hardened more versatility was introduced to the quenching approach and oils or even air where added to the choice of quenching mediums. For the purpose of this writing I will stick to the most commonly used liquid quenchant today- oil.

While, contrary to popular belief, speed alone is certainly not an exclusive qualification for a quenchant, it is important with some simpler steels for the reason of outpacing the aforementioned escape of the carbon atoms from austenite solution. To get an idea of how fast this needs to be, with 1084 pearlite will begin to form at 1000F in around .75 seconds, and it gets down to .5 seconds with 1095. Considering this, one can see how obtaining the most effective cooling from soak temperature to 900F can be critical in fully hardening a piece of steel.

However countering the desirable quality of speed is the need to avoid over stressing the steel during the incredibly intense changes it is undergoing. The internal strain caused by the transformation of hardening is enough to blow the steel apart if it is not kept carefully balanced. Thus the ideal quench for any steel is one that obtains the maximum amount of hardness from trapped carbon with the minimum of stress beyond what is needed to accomplish this.

How it happens

When the steel enters the quench the rapid cooling will take the austenite solution headlong toward a critical point around 1000F where the most stable and favored phase of the iron carbon mix (pearlite) will want to form. This is the most critical point in the hardening operation, for only by keeping the carbon in solution can you get maximum hardness from it at room temperature; every speck of pearlite that manages to form at 1000F means that much less hardened steel at room temperature. If the quench is completely inadequate there will be pearlite and the blade will not harden. If the quench is only partially effective there will be soft spots and less than full hardness as it is riddled with pinpoint pearlite colonies. But if the cooling is fast enough to keep full solution, pearlite will be avoided and austenite will be maintained at temperatures well below its natural range.

bloom.jpg


Above is an image of pearlite colonies that formed in the martensite of old style bloomery steel even when it was quenched in water. Fortunately even our simplest modern steels have alloying to help in avoiding this.

If the latter occurs the steel will undergo no further changes until a point below 500F*, is reached and the need for the iron atoms to assume a room temperature orientation becomes impossible to resist. At this point a new transformation will be initiated; one that is driven by temperature alone with time being irrelevant, where the carbon doesn’t move at all but the iron matrix around it does in a very dramatic way.

The new phase created by this transformation is the hardened form of steel and is known as martensite. From the point at which it begins (designated “Ms” for martensite start), to room temperature, more and more martensite will form and the steel will get increasingly harder. This transformation is dependent on the ability of the matrix to deform and if there is too much carbon, or the alloying is too great, it will resist and you will get retained austenite resulting in lower hardness.

During the strain of this transformation any undue added stress is not only unnecessary, it can be catastrophic, and so it is beneficial for a quenchant to not have great speed in this part of the process. The ideal quenchant will cool very quickly and evenly at first but then decelerate to a slower and gentler cooling below 600F. The reason oils and other mediums were developed to replace water was because water has the inverse of this desired behavior, creating insulating steam at first and then cooling very rapidly at the end of the quench, often resulting in that dreaded “ping” sound.

I could devote an entire separate thread on quench mediums so instead I will merely cover the quenching technique in this writing. An effective quenching method will incorporate a few standard practices which help secure success. Preheating the oil will be one of those practices. Warm oil has a much lower viscosity, greatly increasing its convective cooling. Water based quenchants can be slightly warmed to reduce the thermal shock but the greater problem of closing the gap to the boiling point is another concern. Most oils respond very well when heated from 95F to 130F. Beware the temptation of overheating, as most oils start to lose their conductive cooling rate above 200F, and temperatures in excess of 150F will shorten the life of most quenching oils.

The very short times to make pearlite mentioned above has often been misconstrued as a need to hurriedly rush from the soak to the quench, so much so that I fear people could get hurt. Relax, the .75 seconds for 1084 occurs at 1000F not at soak temperature. Just keep your quench tank near your heat source and promptly quench. Don’t walk across the shop with a blade of 1095 however since the extra carbide will begin to move well before 1000F. But in either case air is a great insulator so the race actually starts when you enter the liquid and you turn matters over to the quenchant, and this is why it so important to have one you can rely on.

Quick and even immersion in the quenchant is important. I have found a cutting motion effective, spine or edge down, for horizontal quenching (blade length held parallel to the liquids surface) and a stabbing motion for vertical quenching. But once in the quench, continued movement between the blade and the liquid is very important to keep fresh liquid available for cooling and break up any vapor films that may form. Industry uses pumps or impellers to move the oil but in the absence of such luxury one can at least move the blade. Many not familiar with the process fear that movement could cause distortion, and certainly improper movement, such as side to side pushing with the flats of the blade could indeed be a problem, but proper agitation will actually eliminate many distortion threats.

Never just drop or toss a blade into the quench, or allow it to rest on the bottom or side of the container. And always remember that it is the vapors not the liquid that burns. Combustion needs three main things to occur- fuel, oxygen and heat, with all hot steel below the surface of the oil it cannot happen, this is important to remember for the long healthy life of both you and your quench oil.

Once the blade can be held comfortably in your hand get it in the temper as soon as possible to relieve some of the stress of the hardening operation, leaving steel in its as-quenched condition for extended periods of time can lead to distortion, micro-fracturing or even all out cracking. More than one knifemaker has found pieces on the bench when waiting until tomorrow to temper.

*the greater the carbon content, the lower the temperature at which hardening occurs will be, many steels with less carbon content than the average knife will begin the transformation above 500F.


Recommended reading on this topic: “Quenching and Martempering” by ASM, “Heat Treater’s Guide” by ASM, “Principles of Heat Treatment” by M.A. Grossman




If this post, and the resulting thread, has information that you feel is useful enough to be linked to in the sticky index at the top of the page you may indicate that by voting in the pole. If you feel the information, or resulting thread, is not helpful, productive, or positive enough for all visitors to benefit from, do feel free to indicate that with a “no” vote in the pole. Both opinions will be regarded with equal value in ensuring only the best information is highlighted in this forum.
 
Last edited:
I have also included below a list of common Quench oils and breif information about them to help understand the properties involved and thier relationship. Also included are links on where to purchse them or find more information:

Medium Speed oils (good for 5160, 52100, O-1, L6 and other alloyed steels)

Gulf Super Quench 70
Viscosity (SUS @ 100F): 81.4, Nickel Ball Speed: 10-12, Time to reach 200C: 41 seconds

Chevron Quenching Oil 70
Viscosity (SUS @ 100F): 82, Nickel Ball Speed: 10.5, Time to reach 200C: 43 seconds
Sources to find it- http://www.chevronlubricants.com/products/metalworking-fluids.aspx

Houghtoquench G
Viscosity (SUS @ 100F): 100, Nickel Ball Speed: 10-12, Time to reach 200C: NA
Sources to find it- http://www.houghtonintl.com/en-us/products/heattreatment/Pages/default.aspx#
http://www.industrial-oil.net/metalworking-quench-oil.html

Park AAA
Viscosity (SUS @ 100F): 85, Nickel Ball Speed: 9-11, Time to reach 200C: 49.62 seconds
Sources to find it- http://www.maximoil.com/

Citgo Quench oil 0510
Viscosity (SUS @ 100F): 58, Nickel Ball Speed: 14.5, Time to reach 200C: 39.5 seconds
Sources to find it- http://hkoil.com/industrial-lubricants/general-lubes#.UNNR6obhdVI
http://www.westernstatesoil.com/IN_MetalOils.html

Fast Speed oils (good for simple carbon steels like 1075, 1080, 1084, 1095, W1 and W2)

Houghtoquench K
Viscosity (SUS @ 100F): 77, Nickel Ball Speed: 7-9, Time to reach 200C: NA
Sources to find it- http://www.houghtonintl.com/en-us/products/heattreatment/Pages/default.aspx#
http://www.industrial-oil.net/metalworking-quench-oil.html
Park #50
Viscosity (SUS @ 100F): 45, Nickel Ball Speed: 7-9, Time to reach 200C: 36 seconds
Sources to find it- http://www.maximoil.com/
 
Great info Kevin, but I would just like to add that in the case of refined vegetable oils quenchants, such as canola oil, there is virtually no vapor jacket stage so they tend to be fast in the initial part of the quench, and slower during the lower temp., part of the quench, naturally. Petroleum based oils exhibit multiple cooling mechanisms with uneven heat transfer coefficients,... requiring cooling rate accelerators, costly additives etc., whereas the vegetable oils don’t and cool by simple direct convection,… and they can make a good alternative to the engineered petroleum based oils in small shop environments where high volume and oxidative stability aren’t as much of a concern.

Another nice thing is that, you can pick them up while you're at the grocery store,... plus they are biodegradable, annually renewable, have generally higher flash points and are non-toxic...
 
Last edited:
I think I read each paragraph twice.
great information...
 
If we are to follow the often cited “Quenchants Derived from Vegetable Oils as Alternatives to Petroleum Oil “ by Lauralice C.F. Canale, Ester Carvalho de Souza, George E. Totten, which was, as the title decribes, a study to see if vegetable oils could be used as a base stock in the manufacture of formulated quenchants, then we must cite it accurately-

“The cooling rates, however, were somewhat faster, especially in the critical martensitic temperature transition range (300°C), than the petroleum-oil quenchants evaluated, suggesting that they may not be as desirable for higher-alloy, crack-sensitive steels. ”

Studies such as this did indeed lead to the development of formulated quenching oils that were an alternative to the skyrocketing price of petroleum bases, and appealed to environmental concerns. Two of which have I been able to find information on, though no real details:

Houghton Bio-Quench 700
From Houghton International - http://www.houghtonintl.com/en-us/pr.../default.aspx#

SoyEasy* Quench Oil
Offered by Sterling Tool and Supply Company http://www.sterlingtoolsupply.com/715/736/index.html

There were many other quenchants I did not include due to lack of information or sources, including these. I would like to have more to add but information is very scarce on these type of quenchants, but I will include what I have found to cover all bases and avoid this thread devolving into yet another quenchant debate, this topic deserves to be covered just once without that happening. Let’s see if it can, shall we?
 
Last edited:
Always a good read and refresher! Thanks for your time and patience Kevin!

You are very welcome. I am sorry that it took so many months between the 1st and 2nd parts of the hardening threads, but business travel and life have been keeping me busy. Then right after I posted it a couple days ago my computer monitor went black, and started whining and producing a burning smell. It took me a couple of days to get to town to buy a new one and catch up on the e-mail and other business.
 
Last edited:
Kevin, would this thread be a good time to discuss why martensite will only form in maximum section of twice the depth of hardening? I accept,for instance, that if I can harden to a depth of 3/32" that I'll only get conversion on the parts of the blade up to 3/16 thick, I don't understand why (am I going to regret asking?) and I'd like to know the likely resulting structure and characteristics of the rest of the workpiece.

Rob!
 
Kevin, would this thread be a good time to discuss why martensite will only form in maximum section of twice the depth of hardening? I accept,for instance, that if I can harden to a depth of 3/32" that I'll only get conversion on the parts of the blade up to 3/16 thick, I don't understand why (am I going to regret asking?) and I'd like to know the likely resulting structure and characteristics of the rest of the workpiece.

Rob!

Sure I would love to discuss it. I am not sure if I follow exactly what your are describing, is that based on a specific formula or what you have observed in your blades? The initial impression would be that you get twice the maximum depth of hardness for a given steel in a blade because of the cooling action from two or more directions. For example, I always get a very high hardness (equaling that of the edge) at the very spine of my blade shaped samples, but around 3/16" in from the spine I will get the lowest readings of all. This is, of course, due to the fact that the quench was able to cool from three directions (both sides and the spine surface).

The vast majority of the microstrucure in the non-martensitic area will be fine pearlite, but there could be some upper bainite or traces of proeutectoid cementite (above .8%C) or proeutectoid ferrite (below .8% C). One cannot underestimate the influence of proper solution (or lack thereof) as well. The more undissolved particles within the austentic matrix, the more places there will be for these diffusional phases to begin forming. Everything we do always seems to be a proper balancing act. Deeper hardening can be facilitated by more thorough solutions at higher temperatures, but so will retained austenite and grain growth:2:. I have achieved acceptable hardness in eutectoid 10XX series steels when quenched into low temp salts by increasing the austenitizing temps enough to facilitate it, but the resulting grain size was not acceptable.
 
Good point Kevin, and there were major concerns about the stability and cost of the vegetable oils in the long run, especially in high volume industrial applications. Canola was looked at particularly because of it’s relatively high stability compared to some of the others.

I haven’t seen anything conclusive about the application of vegetable oil quenchants on “higher-alloy crack sensitive steels”, but they seem to suggest further experimentation and studies are needed. In light of this, I would recommend an interrupted quench and finishing in air. It may also be possible to slow the quenchant down in the initial part of the quench somewhat simply by using it at room temp., or at lower temps., rather than heating it. I also believe the studies were conducted without the use of agitation on the vegetable oils. So,… it seems very possible to speed up or slow them down, without costly additives or by tweaking them chemically.

This should be weighed against the fact that the cooling curve analysis suggests that they are acceptable even for difficult to harden crack sensitive steels.

If I’m not mistaken, the engineered vegetable oil quenchants mentioned are tweaked primarily to improve their oxidative stability, which is fundamentally different than what they do to the accelerated petroleum based quenchants just to get them to work.

http://www.jtcustomknives.com/calendar/files/1/Vegetable Oil Quenchants.pdf

http://www.industrialheating.com/ar...getable-oils-as-alternatives-to-petroleum-oil

http://books.google.com/books?id=io...&resnum=1&ved=0CBMQ6AEwAA#v=onepage&q&f=false

Quote from the third link on canola oil:

* Nearly non-existent vapor phase
* Faster quench rates in the 1300- 1100 F range where high quench rates are necessary to achieve properties
* Slower quench rates at low (900-250F) where low rates are desirable to minimize distortion
 
Last edited:
... It is also interesting to note, that both the vegetable oil quenchants and the petroleum based quenchants exhibit the same cooling mechanism (convection) in the low temp., part of the quench, from what I've been able to gather.

I have found references to non-accelerated and accelerated petroleum based oils, but none to decelerated oils. If anyone has, I’d be interested to hear about that.

Generally speaking, the accelerated oils are used on parts having low susceptibility to cracks caused by internal stress, and I don't think knife blades fall into that catagory.
 
Last edited:
Sure I would love to discuss it. I am not sure if I follow exactly what your are describing,,,,

Less complex that that I think. Let me try again. One would think intuitively, that if you hardened a 1" shpere with a process that gives a shallow hardening steel say a 3/32 depth of hardening - that you would have a hard crunchy shell (martensite) with a soft chewy center. (pearlite?) It has been suggested here before that such is not the case and that martensite will either form completely through a section - or not at all.

This is somewhat in keeping with experience, so I accepted it as explained. By way of example, if I'm hardening 1095 in 1/8, 5/32 or 3/16, it comes out of quench (as one user once said) "harder than a woodpeckers lips!" When the spine or tang is 1/4" or more, I get something more than annealed, but way less than a hard martensite shell. I'd imagined that the tang was probably pearlite - but always wondered why we don't get that hard shell.

Rob!
 
Of the two writings linked to, (two of the three links are reports of the same study with one being a rough “uncorrected” draft, for accuracy I cited the published version) only the Lauralice C.F. Canale, Ester Carvalho de Souza, and George E. Totten writing deals with raw vegetable oils off the shelf, which they hoped to find promise in for use as a base in formulated quenchants. The second article cites the improvement in the later, lower temperature, part of the curve in with a canola based quench oil, as a full, and accurate, quote shows:

“ In cooling curve tests, the canola-based quench oil showed the following characteristics
when compared to a mineral oil based quenchant:

• Nearly non-existent vapor phase:
• Faster quench rates in the 1300-1100°F temperature range, where high quench rates are necessary to achieve properties:
• Slower quench rates at low (900-250°F) where low rates are desirable to minimize distortion”

Knives, would be about as simple in cross section as a heat treated part is going to get, without being a rectangular flat bar. More complex shapes and differential cross sections are what industry is concerned with for accelerated quenches and they get around this most effectively with alloy selections. In comparison the danger of cracking a shape as simple as a knife blade should be minimal with a heat treatment appropriate for the selected alloy.

Tai, I can assure you that challenging you in any way, shape or form was the very last thing that could have possibly crossed my mind the starting this thread. I was specifically asked by the participants of this forum to write these informational threads and I am happy to oblige and feel the people I do them for to be entitled to as much accuracy as possible. And I will be honest that one of the reasons I delayed this thread for months is that I knew some folks wouldn't be albe to let it go without the same tired old squabbling.

I will be happy to advise that a good bottle of Wesson oil will get the job done for any maker not ready to move to an actual quench oil, if it will help get this thread back on track for its intended purpose:

Mazola Cooking Oil
Sources to find it- http://www.walmart.com/ip/Mazola-100-Pure-Canola-Oil-40-oz/19500300?t=1&

So the petroleum base quench oils have been represented in 1 post, and the vegetable based oils have now been represented in 5 posts, I believe we have covered all the bases, with the canola getting more than its fair share. This thread now needs to get back on track in helping with the whole topic of hardening.
 
Less complex that that I think. Let me try again. One would think intuitively, that if you hardened a 1" shpere with a process that gives a shallow hardening steel say a 3/32 depth of hardening - that you would have a hard crunchy shell (martensite) with a soft chewy center. (pearlite?) It has been suggested here before that such is not the case and that martensite will either form completely through a section - or not at all.

This is somewhat in keeping with experience, so I accepted it as explained. By way of example, if I'm hardening 1095 in 1/8, 5/32 or 3/16, it comes out of quench (as one user once said) "harder than a woodpeckers lips!" When the spine or tang is 1/4" or more, I get something more than annealed, but way less than a hard martensite shell. I'd imagined that the tang was probably pearlite - but always wondered why we don't get that hard shell.

Rob!

Oh! I get it now. Actually in large cross sections you could only get an outer shell of hardened steel if the quench is fast enough to overcome the thermal extraction via conduction of the metal. That is how the Jominy quench test works, by directing a water spray at the outer surface of the steel and measuring the drop of hardness as you move away from that outer surface. My quench test samples are typically 1095 that are a 1-1/8” wedge which is 1/6” at the edge and 5/16” at the spine. They are fully quenched the then cross sectioned so that the metallography and Rockwell readings can be done at the core. The hardness readings are taken at approximately 1.5mm apart, from edge to spine after micrographs are taken at similar intervals to give corresponding microstructures for the readings. In these size of cross sections water based quenchants are more than capable of just through hardening the stuff, but with various speeds in oils you will get different effects of penetration of hardness. I don’t have many numbers for the travel from side to side on the cross sections since the edge to spine is the critical part for a blade, but your question will prompt me to take a few more in the future. I do have metallography in those areas and with pearlite there will be wider lamellae spacing nearer the core and some of the martensitic shell effect.

You may find clues to your answers by studying what is known about thermal transfer rates for steel when it is heated. If you have, or can get, a copy of “Tool Steel Simplified” by Palmer and Luerssen, there is an excellent chapter on “the time required to heat tool steel” in which they dispel a long standing myth that one can have the center of a large section “cool” while the outside has assumed soak temperature. In fact the conductivity of the steel will prevent the outer surface from assuming the furnace atmosphere before the center because the center will act very effectively as a heat sink until all areas equalize. “How does that explain and edge getting hot before the spine?” some may ask. Well, that is a matter of an entirely different cross section not part of the same cross sectional thickness. Imagine a part with many little splines extending from it, like a gear, this would be one of those parts, discussed in the last post, that industry is concerned about with accelerated quenches because those splines can heat and cool independently from the main part and then distort or even crack. However Your sphere analogy would be the simplest possible cross section and would not be able to see those kinds of heating gradients.
 
Kevin, I'm wondering if he is talking about what Verhoeven called the film heat-transfer control that he mentioned in Steel Metallurgy for the Non-Metallurgist that explains why, with a shallow hardening enough steel that you will get a martensetic edge and a pearletic spine. Rob, if that's the case, when you get down to cross sections less than around 1/4" inch there is little difference between the core and the surface of the steel when it comes to heating and cooling. As Kevin mentioned above, when quenching a thicker piece of a shallow hardening steel the mass will keep it from cooling rapidly enough to form martensite before it completes conversion to pearlite. Where the steel is thinner, as towards the edge, where the steel is no more than twice the thickness of the depth of hardening or less, the steel will form martensite. This does, of course depend on the thickness of the steel, the alloy, and the grain size. The speed of the quenchant will be in there too, I imagine. If the steel is thicker than that then other forces come into play as Kevin mentioned.

I hope you don't mind me chiming in and I'll let Kevin straighten out anything that I've misstated.

Doug
 
Yes Doug. You've often made referenece to martensite only to twice the depth of hardening. If I'm reading you both correctly, the answer is simply heat sink effect keeping even the outer parts of the section from getting under the pearlite nose. Whoda thunk? :biggrin:
 
Keven, I have a weird question. Is it possible to use fast quench oil like parks 50 for all steels, say O-1? Could doing an interrupted quench provide the correct timing for the quench to allow for O-1 to quench properly?
 
Keven, I have a weird question. Is it possible to use fast quench oil like parks 50 for all steels, say O-1? Could doing an interrupted quench provide the correct timing for the quench to allow for O-1 to quench properly?

Hmmm, that is a tough one, and certainly not the first time the thought has been entertained. Technically, and theoretically, one could. But they would need to nail the timing on the interrupt just right, and hopefully every single time. I make the assumption that if somebody has taken the step to get a good quench oil that they are seeking greater consistency in their results, this route would necessitate reintroducing a pesky variable that you would have to stay on top of every time. Sort of like getting a calibrated oven only to clip all the thermocouple leads in order to open the door every minute to still do a magnet check. I do interrupted quenching when I have to, but I much prefer actual marquenching instead.
 
Originally Posted by theWeatherman
Keven, I have a weird question. Is it possible to use fast quench oil like parks 50 for all steels, say O-1? Could doing an interrupted quench provide the correct timing for the quench to allow for O-1 to quench properly?

Hmmm, that is a tough one, and certainly not the first time the thought has been entertained. Technically, and theoretically, one could. But they would need to nail the timing on the interrupt just right, and hopefully every single time. I make the assumption that if somebody has taken the step to get a good quench oil that they are seeking greater consistency in their results, this route would necessitate reintroducing a pesky variable that you would have to stay on top of every time. Sort of like getting a calibrated oven only to clip all the thermocouple leads in order to open the door every minute to still do a magnet check. I do interrupted quenching when I have to, but I much prefer actual marquenching instead.


I bet this thread could go on forever! I have been using P50 for my O1 neckers, 7/64 by 6" by 1", and have not had an issue with cracks that I can see and I have broken them in half in the vice and they did not break any quicker, meaning less degree of angle from 90*, than my 1084. The edges hold up very well, past the brass rod flex and so in my limited knowledge I believed I acquired a successful ht and temper.

Is using P50 for larger blades in O1 the issue due to the larger thermal mass? If so then could you define what a larger blade is or what you would consider one to be?
 
Last edited:
Shane, you bring up a good point that works into the many of the topics involving the quench, that of mass. Due to metallurgical testing I have the benefit of seeing the effects of a wider array of cross sections than most knifemakers do. As I have mentioned previously we are blessed with incredibly simple cross sections when working with knife shapes and the thinness of them saves us more grief than we realize. I have found that a steel like O-1 will tolerate “over-quenching”, for lack of a better word, much better in ¼” or bellow than in thicker sections. For example a ¾” thick section of O-1 will be much more prone to cracking when quenched in #50. The real killer is a thin section attached to a thick section as the expansion rates will be so differential that something will have to give. I tend to quench many metallography samples in the #50 just to insure that I get complete conversion, but sometimes there will be micro-fractures. Since I do not temper samples that are intended for pure martensite study, almost all of the standard 10mm x 10mm samples I use will exhibit major fractures within two weeks if not tempered. Blade shaped cross sections tend to stand up better. It would appear that not only are simple blade shapes a bit more tolerant to less than ideal quenching, they are also rather stable shapes for handling residual stress issues as well.

Expansion rates are key to much of this, because I also just thought of a situation where a blade cross section will not be able to cope at all. Folks who have done a san mai constructing with less than compatible steels have seen this. The worst offender I have seen yet is a thick slab of 52100 down the center of some other odd damascus mix that is then slightly over-quenched. I have seen such blades literally pull themselves into two pieces almost from tip to ricasso with the crack running down the center of the 52100.

When steel hardens at, or below, Ms (500F - 400F approx.) the expansion is VERY dramatic. If that expansion is not even the steel will distort to accommodate that, if the steel is too strong (via hardness or thickness) to accommodate it, then it will come apart.
 
Last edited:
Back
Top