Working the three steel types.

[Kevin R. Cashen]

I have taken the liberty of compiling the two stickies (hypereutectoid and eutectoid) together within a cleaned up version that only includes the on topic posts, minus the kind positive one liners as well as other posts. I have also added the final chapter that was always missing on hypoeutectoid steels. It was quite a bit of work to do but I would like to give it all as a gift to the forum to replace the original two threads in the crowded sticky area. I hope I have broken no rules nor offended any with this action, and if it is not appropriate you can allow this new thread to simply slip off the front page and fade away as the natural flow of the forum takes it course, or if it is acceptable repalce the other stickies. I hope they will be enjoyed by all as much as I have enjoyed my time at this forum.

I apologize to anybody whose posts got dropped in trimming it down, I did the best I could with quotations, and no slight was intended, but if one wishes to see the initial threads in their entirety they are here:

http://www.bladeforums.com/forums/showthread.php?t=615086

http://www.bladeforums.com/forums/showthread.php?t=615784

To help in using all this information I have also included heat treating specs on my web site for most the the common steels. On the "hardening" page, and other spots, simply clicking on the steel name in the list will take you there: http://www.cashenblades.com/info/hardening.html

 

[Kevin R. Cashen]

Hypoeutectoid Steel

In the posts that follow we have talked about hypereutectoid steels (>.8% C), and eutectoid steels (around .8%C), which only leaves hypoeutectoid steels (<.8%C) yet to touch on. Air cool a hypereutectoid from above critical and you will get maximum pearlite with left over carbon (cementite). Air cool a eutectoid and you will get little else than pure pearlite. But when you heat a hypoeutectoid steel above critical and slow cool you will get incomplete pearlite within an excess of iron (ferrite).

As the carbon level decreases below .8% the more ferrite you will have that is not infiltrated with carbon. Think of iron as bread and the carbon as butter in a little dab in the corner. The natural state of the butter is in this little dab and when given a chance it will unspread itself and return to this dab. Heat is your butter knife, the more heat you use the more you can stretch that little dab of butter to cover the entire slice of bread. This is why you will find recommended austenitizing heats increasing with lower carbon contents.

Because steel requires a minimum of carbon to fully harden, hypoeutectoids have a narrower range that is useful to knifemakers than hypereutectoid steels. One can get maximum quench hardness (65-67HRC) from a simple carbon steel at or above .8% C. From .8% down to .6% the maximum quench hardness decreases rather slowly, but below .6% maximum hardness drops much more rapidly. So for edge holding the hypoeutectoid range of .6% to .8% carbon is the most desirable. Large choppers that rely more on cleaving type cuts can be made with steels from .6%C down to .5% C but strength issues will become a factor in lower ranges, requiring thicker and more obtuse edge angles to compensate, think of axes and hatchets instead of knives.

Hypoeutectoid steels have greater toughness just from the lower carbon content reducing the maximum hardness, but although there is less inherent risk of brittleness there is also less strength from higher hardness levels. Fortunately there is another factor that naturally gives these steels an edge on toughness; lathe martensite. There are two common morphologies of martensite referred to as &#8220;lathe&#8221; and &#8220;plate&#8221;. The form created is actually dependant upon Ms temperature, however since Ms is determined by chemistry we can look at it from a carbon content perspective. Below .6% carbon (notice reoccurring patterns here?) a steel will most likely have all lathe martensite, above 1% carbon the steel will have primarily plate martensite. From .6% to 1% (the range we knifemakers love) there will be a mixture of both. Plate martensite is rather chaotic and its units tend to impinge on each other at high angles resulting in strain issues that render it more brittle, sometimes to the point of micro-fracturing. Lathe martensite forms in orderly little packets that nestle next to each other more comfortably, making it naturally tougher.

So in heat treating, hypoeutectoid steels respond better to higher temperatures. For annealing one can forget all the hassles of spheroidizing if they wish and go with the old vermiculite or wood ash thing, there just isn&#8217;t enough carbon in most cases to cause issues. In fact industry prefers not to spheroidize hypoeutectoid steels because it makes them &#8220;gummy&#8221; in machining operations. In tempering be certain to start low and carefully work higher as I have found hypoeutectoid steels to drop in hardness quicker than their carbon rich counterparts.

I personally do not work much with hypoeutectoid steels as I enjoy the strength imparted by higher carbon levels which can still be managed quite well with alloying for toughness. I also find the Ms points to be not worth the hassle with the marquenching techniques I prefer. But if you do work with them and want to marquench, be particularly aware of the fact that Ms starts at increasingly higher temperatures as the carbon levels drop. Why is this a concern? Let&#8217;s examine a typical marquenching operation with 1050. If you go for Ms as your interrupt you will be air cooling from just under 600F down to ambient with M-90 (90% martensite formed) above 475F. The maximum hardness is already only 60HRC but the auto-tempering will occur entirely above 475F with most of it having the effects of a 500F+ temper. Final hardness would be lucky to reach 52 HRC.

If however you stay with a set interrupt of a lower temperature that would work for higher carbon steel, say 400F., there really isn&#8217;t any point in marquenching at all since you will be stopping the quench well below M-90; which is actually just a normal quench.

Heat treating information for a common hypoeutectoid steel (5160) can be found on my site here: http://www.cashenblades.com/info/steel/5160.html

 

[Kevin R. Cashen]

Eutectoid steel

We so often recommend 1084 or 1080 to makers who are getting started why not explain why that is? In the other thread I avoided including the iron-carbon equilibrium diagram to stay away from making things too technical, but it turns out that it would have been beneficial in understanding many of the topics discussed. So to get a better idea of this topic I will go ahead and put his here:

I am not sure where I originally got this image but it has been tucked away on my site for some time after some discussion in the past.

The iron-carbon equilibrium diagram charts out how iron and carbon will combine under different carbon concentrations and temperatures. Along the bottom you will find from left to right increasing percentages of carbon. Along the left side you will find increasing temperature from bottom to top. Virtually everything we are concerned with in knife steels will fall well to the 2% mark that divides steel from cast irons, and the vast majority of simple knife steels will fall to the left of the 1% carbon mark. The only thing worth mentioning that is to the right of the 2% mark is the &#8220;V&#8221; shaped line that occurs at around 4.3% carbon and 2000F this deals with the eutectic and I only point it out in order to remind you to never confuse it with the eutectoid. The eutectic deals with actually moving from liquid metal to solid, the eutectoid occurs entirely within solid metal but deals with solid solutions with that metal.

Back to steels, and back the thread topic. At .83% carbon you will notice a very prominent dotted line ascending all the way up and dividing the entire range of steels almost neatly in half. This is the eutectoid. Anything to the right is hyper-eutectoid; anything to the left is called hypo-eutectoid. When steels to the left are slow cooled they will have some pearlite with leftover ferrite (iron) that was not filled with carbon to make that pearlite. When steels to the right of that line are slow cooled they will be pearlitic with leftover carbon that didn&#8217;t get used. Right on the line the steel will go entirely pearlitic with no leftovers at all.

If you pick a point in the hypereutectoid zone, let&#8217;s say a steel of 1% carbon, and make an imaginary line straight up like the dotted line at the eutectoid you will be tracing things that happen to that steel with increasing temperature. The first line you will cross will be A1 at 1333F, since we are going up in temp we should actually add a &#8220;c&#8221; and call it Ac1, I could explain all of this but it is a bit involved and actually involves French terminology that is not really necessary for this thread. A1 is the point where the steel will begin to move carbon around and start going into solution. The point of greatest solubility of carbon into iron at this temp is .83% so that much carbon will readily go into solution at this temperature, but .17% carbon of our 1% will still be left undissolved. Increasing the heat will result in more an more of that carbon going into solution until you encounter the line that rises on an angle from the eutectoid line up to the right and is labeled Acm. This is the line that show when you have dissolved all that extra carbon and put in into solution. On the left hand side of the eutectoid you will see a line that is very similar to Acm but is called A3. A3 is where all the iron (ferrite) has been filled with the carbon that is below the eutectoid. It is also worth noting the line labeled A2 this is the Currie point or where the steel loses magnetism at 1414F. It is VERY important to take not that this A2 line is flat a horizontal, and thus not affected by the carbon differences like A3 or Acm. This is exactly why it cannot be stressed enough that the magnet is NOT an absolute or foolproof method of determining heat treat for every alloy.

Now back to our thread topic. Right in the center of these two lines (A3 and Acm) is the eutectoid point. A steel falling near this line will make total pearlite on slow cooling with no problematic leftovers, and on heating will go totally into solution at the lowest temperature of any steel. Because of this if all you have is a forge or a torch 1080 or 1084 will cooperate with you more readily than any other steel. Simple heating tools will make it much more difficult to pinpoint and hold temperature between A1 and A3 or Acm in order to avoid the problems that those extra components can cause. The safest way to get the best results with these tools is to use 1080 or 1084, but if you are going to work with steels to the left or particularly to the right a very good knowledge of heat treatments that can be done prior to hardening to set things up can be invaluable.

Heat treating and other working information on a classic eutectoid steel can be found at my website here: http://www.cashenblades.com/info/steel/1084.html

Here is a link to a chart of ferrite/pearlite/cementite in steel and a Carbon-Iron Equilibrium Diagram (pages 4 & 5) that may or may not be helpful...
www.msmw.com/Basic%20heat-treating%20of%20Carbon%20Steel.pdf Page 4 & 5.
Mike

Please discuss why one still wants to take eutectoid steel above A1 in order to heat treat properly. Since the diagram is for "neat" iron and carbon, what effect if any does the presence of the manganese have on the solubility and mobility of the carbon?
Finally, since A1 and Acm are essentially at the same point for the eutectoid, does this indicate that it is much more sensitive to grain growth?
Thank you.

mike Fitzgerald

A1 - you must transform the steel to austenite to HT -that means above A1.
I have to assume manganese has little effect on carbon solubility or diffusion.[I'd have to research to be sure]. Remember that carbon is much smaller than most of the alloying elements in steel ,it can move easily between other atoms and that's where it stays in the crystals.The big effect of Mn is in hardenability.
The eutectoid might be a bit more sensitive to grain growth as the carbides in the hypereutectoid would tend to slow down grain boundary movement. Nothing that a pinch of vanadium wouldn't cure, but that would help all of them !!

The reason I asked my question was to clarify this statement of Kevin's:
"A1 is the point where the steel will begin to move carbon around and start going into solution. The point of greatest solubility of carbon into iron at this temp is .83% so that much carbon will readily go into solution at this temperature"

I understand from HT data that it needs to be heated above Ac1, but since Acm and Ac1 are the same point on the graph at the eutectoid, how do we deduce where the proper temp is for converting to austenite?

mike Fitzgerald

Theoretically Acm and A1 are the same point for .83% carbon and iron. The problems arise when reality enters the picture.
The transfer of heat is not a 100% situation, nor is it instantaneous. In an analytical oven with good insulation, and good regulation, Eutectiod steel could convert to Austenite at 1 degree above A1/Acm.In reality, it would take a very long time for the solution to fully occur at the exact point. Most charts will add 50F to assure a full and reasonably fast conversion.

Most new smiths have basic and primative equipment, no control ,and are guessing the temperature. Thus, giving the temperatures in a higher range and allowing a short soak are designed to prevent failures by erring on the side of caution.

I am glad that Kevin posted this. I almost always recommend 1080/1084 as a good starter steel. The reasons are usually lost on the newbies, and augured by some of the others. The very reasons that make 1080/1084 a good starter steel make 1095 a much less desirable starter steel.

Stacy

 

[Kevin R. Cashen]

Thanks, Stacy, that's a nice pragmatic explanation but I asked rhetorically so that perhaps we could get a little deeper into it. We know that the carbon starts moving around at Ac1. But we also know that there is an actual solid state phase change that it has to undergo to become austenite. This is an equilibrium diagram, steady-state, so it is understood to be unlike real-world conditions.
My question is more along the lines of what is the next step in the transition after initiating carbon movement? What is transpiring at the point it becomes non-magnetic but still too low a temp for full austenitization? Is it a gradual transformation to austenite over the next 50*, or rapid and global once a certain energy threshold is achieved? What is the difference of whether it starts out as martensite versus pearlite?
Since we hopefully have an opportunity to have a thread go a little deeper than has been typical recently, I hope to stimulate Kevin and mete to give us a dose of the heavier stuff.

mike fitzgerald

Good question Fitzo, so good that it has to be answered carefully since it would be very easy to spin off into some very technical discussions involving the physics involving energy equilibrium in grain boundary movement. But to keep it simple, there are at least three things working for you on controlling grain growth in a steel that goes into solution at Ac1. First you still have diffusion time at temp working for you which brings us back to our friend the soak time. Next there is the fringe benefit of the aluminum killing process that was used to deoxidize the steel resulting in aluminum nitrides stabilizing the grain boundaries, most modern poured steels are would have this and are actually used to be referred to as “fine grained” steel as opposed to the silicon killed steels that would grow grains much sooner, yet another HUGE difference between the steel the ancients worked with and ours. These particles will hold the grain boundaries in place until enough temperature is thrown at them to defeat the effect and then the boundaries will move very quickly like rubber band that suddenly got freed, this is why grain growth surprises us so much when it happens and gets out of hand so quickly. Also manganese while not forming stable carbides in the boundaries will still have an effect as a substitutional atom in the matrix and slowing the movement of the grain boundaries as well (i.e. solute drag).

Another thing to remember about grain growth is that it is all about transfers of energy and when that energy levels off the grains boundaries tend to become stable until more energy is introduced, this is why some steels will tend toward a particular ASTM grain size. When I consider this it brings to mind yet another possible advantage of a controlled soak, Uneven grain size can be a problem and time at a set temperature could give the grains the opportunity to go from a mixed grain size to a more uniform one, but you need to be sure that the larger grains are still a decent size since that is what things will more toward.


As for desired temp above Ac1, well when you consider the previous paragraphs and how much stability in grain size you really can have, then you can see why a eutectoid steel is the most heat treat friendly of them all. If all you have is a magnet and a forge, a bit above 1414F (Ac2 or the "Currie point") will still work fine for you, if you have the ability to hold at 1500F even better, and you can assure total solution without any of the headaches that can arise with the hypereutectoid steel. Simply put all that carbon into solution and quench away!

Now is a great time for another plug for Aldo and his 1084

Thank you for the explanation, Kevin. It helps explain why we can hold even a simple steel at a given temp for an "extended" period of time (with good control) and not experience runaway grain growth. Phillip established this for us with his experiments with a tool steel. Your explanation helps to understand how it extends to plain steel, also, with trace levels of stuff from the smelting process. Still, I understand the mechanism is complex, as smaller grains will have to continue to grow faster than the large ones if the size tends to homogenize.

I appreciate you trying to answer technical questions in detail but in terms we can all understand. It is enjoyable trying to understand and picture what is going on in that steel at the microscopic level as it is having huge amounts of energy pumped into it.

mike Fitzgerald

So if I understant all of this grain growth, then it is more a product of temperature than a product of time. HIgher the temp more growth, right. We can hold a metal, say 1084, at 1500 all day long and not really expect any growth in the grain.

Wade, yes ! That's why we speak of temperature so much more than time .For those who HT with a torch we warn about not putting the torch near the edge and especially near the tip. The tip can very easily be overheated and the grains can grow rapidly even in short times....If you're a math type - temperature is usually a linear function and time a logrithmic function.Look at a TTT curve ,it's demonstrated there.

 

[Kevin R. Cashen]

Kevin,

At which temperature or which line that the nucleation starts on the heating cycle.

There appears to still be some discussion of where the eutectoid composition lies. I've seen it as low as 0.77% C and Kevins graph is the highest I've seen. Anyone know if its been nailed down? My books are from the late 90's. Until seeing Kevin's chart, I thought 1075/1080 was the closest to the eutectoid composition.

Hey, that diagram is in my materials science book!

The teacher told us all to look at it, make sense of it over the week-end, and come back in ready to discuss it. I must talk too much like I know what I'm doing, because several kids in class wanted me to break it down for them.

me2, they'll never nail it down! Just consider 1080 +-5 . I think Kevin has a book where they give different compositions in the same book ! Don't worry about it because that's the equilibrium diagram of just iron and carbon.Steels you use will differ.

I have couple of books that on one page shows the eutectoid at .77% and on the very next page assigns it at .8%. the highest I have seen it assigned is .85%. But then it all depends on even the slightest of alloying, even in the trace ammounts, the diagram is based in pure iron and carbon at equilibrium conditions, Murphy irrefutably established that perfect conditions in the real world are incredibly unlikely. A case could be made to explain the alloying effects in the way the earlier books tend to place the eutectoid at the lower end than newer texts, perhaps reflecting and increase in trace elements and alloying in all steels.

If I had to assign one point on the diagram to the point of the initiation of new crystalline formation it would be Ac1, but that would be a rather inaccurate oversimplification. In the formation of new austenite grains there are several steps: recovery, recrystallization, and grain growth. Recovery starts well below Ac1 and is the process behind proper stress relieving heat treatments which are entirely subcritical in temperature. Recrystallization cannot be assigned a fixed temperature since it is accomplished over a range and is affected by many factors (residual strain energy, rate of heating etc…), but the conditions that promote it are definitely under way at Ac1. “nucleation” the process that will determine the point of the new embryonic grains actually has its origins in the recovery phase when very tiny subgrains can begin to from in areas of high energy of the current grain boundaries.

Pure carbon/iron equilibrium is 0.80% C at eutectoid? Other elements and amounts of elements cause eutectoid to be higher or lower than 0.80% carbon? Like... if iron is in a mix with other elements, the carbon amount has to change to be eutectoid?

This is "the real" question... If a specific batch-chemistry of steel has a number of elements besides carbon/iron and it is eutectoid at 0.77% C, is the same percentage mix of other-than-iron elements with a carbon amount of 0.83%-0.85% hypereutectoid in the sense of excess carbon that can go to grain boundries?

Mike

You want the real, real, real answer ? If the eutectoid is .77 C then any more carbon can then be a problem.Depending on what you do with it it will appear in the final knife as carbide spheres or in the grain boundary. Much of the point of this thread is to be aware of the carbides in hypereutectoid steels and learn how to make them do what you want them to.
Understand grain boundaries .There is physically more room in them and they have higher energy levels. That means things tend to collect in them ,good and bad, and things happen there , transformations etc. Carbides and phosphorous can collect there resulting in brittleness.Vanadium collects there to retard grain boundary movement thus grain growth.

I was figuring most steel is fully killed but didn't know for sure... I've already got a world's worth of teeth marks on my butt from "figuring"...

Here's a link to a 3/1/09 BF thread with further info on eutectoid/hypereutectoid steel... http://www.bladeforums.com/forums/sh...99#post6566699

Mike

 

[Kevin R. Cashen]

Im new to the world of starting from scratch.Ive made knives before but never from a total blank steel.I would like some advice on which steel to start with.Im wanting to make some Tactical/survival knives,full tang with a para cord wrap handle. Id like something not to expensive(ya know for practice sake) and realitive available.Also somthing not to tough to work with.Any Ideas? thanks,Jeff.

I would trade any of the stickies with my name on them for one good one by anybody on the topic of this question. In the line up of most asked question that could wear out the most determined veteran forumite this is very near the top. But as is so often the case, it is not the person asking the question that makes the topic difficult, it is all the confusing answers that keep resurfacing as if the question was never asked or discussed before. One thing that would help with this topic is to remember the question itself. We are not being asked what our favorite steel is, we are not being asked what the new steel in vogue this month is, and we are not being asked what’s the best super-duper indestructible forever cutting steel is. The question is- what is a good steel for beginners to get started with; an easy steel for anybody to learn on without huge investment in tools or materials. If we are going to get down on free scrap steel we should probably have an alternative answer, and on that point I am tickled to death and almost giddy to be able to help out a new guy who is on a budget but still wants to know about a real steel and not how to make a knife out of mystery scrap they found. Jeff, I applaud your approach and would honor any request for information you may have based upon that position which allows us a good starting point from which to help you.

The easiest steel to get optimum results with little effort or equipment first and foremost must itself be simple! If you plan on doing shade tree mechanics work, a Ford or Chevy from the 60’s or the 70’s is going to be much more rewarding than a 2009 Ferrari or Formula 1 car, which you will probably just foul up without the proper experience and tools.

In steel there are two things at odds in easy heat treatment- how it has to be heated, and how it has to be cooled. Since it appears that most bladesmiths have enormous difficulty wrapping their minds around effective quenching concepts, too many in the business have focused far too heavily on steels that will “appear” to easily harden in any quenchant (I won’t even get into how this can immediately be solved by quenchant choices ).

Alloying is what makes the steel respond to a wider range of cooling in hardening. Two elements are most used for increasing hardenability- chromium and manganese. Of the two, chromium is the more powerful but it also forms carbides and makes the heating more critical. Manganese greatly increases hardenability but does not give the same issues in heating. So in the big picture the element that will allow you to harden in more quenchants and still allow a carefree heating is manganese. I could go into carbide formers and edge retention but, once again, we are talking about a beginners steel choice, so let’s go for skating that file and cutting a few things and worry about the ultimate super knife that can cut through granite and still shave for a few months later in our career.

The next thing we have to worry about is carbon content, too little and the beginner will have a bit of sorting to do to figure out why that file doesn’t skate. Too much and there can be some serious annealing issues and embrittlement problems as well as steel that is much less tolerant of overheating which is a common problem for beginners. Less carbon will increase soak temp requirements to put excess iron into play, more carbon will increase soak time to break the carbide groupings up and put them into play- particularly, once again, if there are any carbide forming alloy elements- like chromium. What we want is a Goldilocks zone for carbon, not too much, not too little, but just right. That Goldilocks zone for iron/carbon systems is around .80%, and is called the eutectoid. It takes the least heat and time to go happily into solution without any leftover material to cause you problems.

So let me see, alloys that are really simple and have around .80% carbon and manganese to help in hardening… perhaps 1070, 1075, 1080, 1084. Hmmm, yes they seem to fit the bill!

5160 is hypoeutectoid and relies on chromium just to be able to reach 62 HRC. It is hard to make a bad knife out of it but it is also hard to make a really great knife out of it without very special attentions, indeed it excels in mediocrity in the hands of a beginner.

And then we have the old favorite on the list of beginners steels, O-1. Why is it believed to be so good for beginners? Because you can quench it in anything cooler than an oven or forge and expect to see some gains in hardness, but the soak time and precise temperature requirements in dealing with all the alloying present make it so the beginner is only getting a fraction of the steels potential at over 3 times the price!

With all of its proeutectoid carbon wandering around with no guidance to keep it out of mischief, and its significantly lower Mn content 1095 is not too much better than O-1 for a beginner. It will challenge your quench speed abilities as well as your heating skills.

Lowest price- 1070, 1075, 1080, 1084
Easiest accuracy and time requirements in heating- 1070, 1075, 1080, 1084
Hardenability in any decent quenchant matched to is cooling needs- 1070, 1075, 1080, 1084
Easy to temper for excellent edge holding or high toughness- 1070, 1075, 1080, 1084
Easy to anneal by simple heating and cooling or full lamellar via insulative methods- 1070, 1075, 1080, 1084
Easy to grind, file, drill or forge-1070, 1075, 1080, 1084
Readily available most anywhere- 1070, 1075, 1080, 1084

So what may be some of the best steels for a beginner to start with, hmm let’s see… perhaps 1070, 1075, 1080, 1084.


P.S. Going back to the concept of addressing the question of "best for beginner" versus "our pick for the best steel ever", I would like to go on the record as saying that I rarely work with 1070, 1075, 1080 or 1084. Not because they are bad steels, they are really good steels, but I have spent more than a few years of my life studying steels and developing experience and equipment that can unlock the potential of other steels that a beginner would probably waste their money on. Thus the steels I suggest for this topic are rarely the same that I choose to use or would suggest for anybody with years of experience and a well equipped shop.

 

[Kevin R. Cashen]

Can 1084 be air quenched, unlike 1095 which has to be oil quenched?
I've been wondering why folks recommend 1084 over 1095 for most carbon steels. Do they grind different too??

Hope ya don't mind me joining in Dixie, but I'm after the same answers you are, I believe.

All of the 10XX steels are shallow hardening, none will do anything but get soft by air cooling, as all they will do is form pearlite. All are technically "water hardening" steels in thicker sections due to their lower hardenability. So I would prefer not to call 1084 a deeper hardening steel since it is one of the shallowest hardening, just 1095 is even more shallow hardening due to less Mn. It is kind of like saying a tortoise is a fast animal simply because it beat a snail in a foot race.

1095 has both the whammies- trickier in the heating as well as in the cooling. 1084 is ready to quench as soon as you get it evenly to the popular non-magnetic standard. 1084 will just want to make 100% fine pearlite on every heating a cooling cycle so it is easy to keep it very homogeneous and even on the inside from forging to final quenching. 1095 on the other hand has carbides to deal with, albeit simple ones but carbide all the same. It can set itself up in the forging, normalizing and annealing stages with some very obnoxious issues that could still be there during and after quenching. It takes a little bit more of a knowledge and skill set to effectively deal with these quirks. Because of this we have gotten some of the unorthodox heat treatments that bladesmiths come up with, or the steel gets an undeserved bad rap because its particular needs were not met.

On one hand, I lucked into this, I think it's fantastic, I've had extremely good results with 1084 (except the few times I managed to crack it messing with brine). I've done a little with 1060, 1070, 1075, but mostly Al's 1084. It really is a good starter material and you can easily learn heat treat variations for other steels by way of their differences from 1084. (hold times, temperatures, etc.)

On the other, 15N20 wouldn't have ever been on the list- and I love that steel. I'd put it on any list of great beginners' steels if it was more widely available, especially in 1/8 inch thicknesses. The availability makes it not work out for the list, though (unless you are crazy enough to use 3/32 blades to baton through logs and then sell knives based on that usability.)

You are correct and if it were not for the size and availability issues I would put 15n20 on the list as well since one can consider it 1075 with some nickel added. It is worth noting however that 1084 should be a better edge holder than 15n20. In my steel selection categories 15n20 falls under tough chopping and cleaving type blades and not under fine slicing and edge holding, but it is a serious problem to make a beefy bowie out of shim stock .

 

[Kevin R. Cashen]

Hypereutectoid steel

I thought I would start a new thread for this since similar questions have been asked in at least two threads and in some e-mails I have received.

First lets explain some of the words that although can have as many as 5 syllables really are not all that intimidating or big. I will not get as involved to post an image of the iron-carbon equilibrium diagram, if somebody else wants to they can, but instead I will just quickly give some details on what I am talking about.

Due to the way carbon interacts with iron in steel we get different phases with the metal depending on how we cool it from solution. At around .80% carbon the two will form pearlite as the carbon separates from the iron. Also in this range the steel will level off in hardness when quenched. Beginning at .60% carbon maximum quench hardness will begin to level off and at .80% adding more carbon alone will no longer get you any more in hardness. When you look at this along with the fact that .80% will make total pearlite with no leftover iron or carbon, you can see that this range is very noteworthy in looking at how we treat the steel.

This range is called the &#8220;eutectoid&#8221;, be careful not to confuse it with the eutectic, which is an entirely separate thing and not he same. The eutectoid is basically the carbon content that is most efficient or simplistic in its use of carbon and the temperatures to drive those changes.

Steel with carbon contents in excess of the eutectoid are called &#8220;hyper-eutectoid&#8221;, and require a little more thought and car in their heat treatment than eutectoid steel. When you have all this carbon .80% of it will be used up in the making of the desired phases, but then there will be leftover carbon to deal with. This leftover carbon can be useful or can be a great liability depending on where you put it.

Grain boundary carbide: When you heat a hypereutectoid steel high enough to put all that carbon into solution, upon cooling it will want to come out of solution, how it comes out can give you problems. Since anything above the eutectoid is not in equilibrium at lower temperatures it will want to come out first, and it will form separate iron-carbide, but where it forms is the problem. If you cool slowly enough from solution the carbide will form in the grain boundaries, and since the vast majority of fractures in hardened knife steels travel along the grain boundaries you have now filled that path with the most brittle material you can. No matter how soft the grains themselves are, the path of fracture is still brittle. After that extra carbon comes out of solution, the next thing to happen is at 1000F when the rest will separate out in alternating bands with iron forming pearlite. Grain boundary cementite is rather stable and will take much more to put back into solution and remove, so once it is there it can survive through the final heat treatment an into the finished blade. I have seem many more whacky bladesmith heat treats dealing with hypereutectoid steel, leading me to believe that much of it could be a matter of dealing with previous heat treat shortcomings that lead to carbide issues.

Retained austenite: In quenching a hypereutectoid steel that has that extra carbon in solution problems will arise when it comes time for the transformation to the hard phase of martensite to occur. All of that carbon trapped in solution will tend to reinforce and stabilize the current phase of austenite, making the transformation. much more difficult. This will result in retained austenite and overall lower hardness levels. I have done tests where one could actually watch hardness drop in a hypereutectoid as I increased the soaking temperature. 52100 are very prone to this and one can watch the HRC drop with any significant increase in temperature that brings more carbon into solution. Once you reach room temperature that is it, the quench is done and you have what you have, this is where cold treatments on steel come in. Extreme under cooling will provide the incentive for this transformation to complete, but freezing steel is no substitute for going things correctly in the first place.

Plate martensite: There are two morphologies of the phase that is hardened steel. There is lathe martensite and plate martensite. Lathe martensite forms at higher temperatures, while the plate morphology forms at lower temperatures and thus both are mostly dependent on the chemistry which determines these start points. The major feature in that chemistry is carbon content, which you the heat treater determine with your soak time and temperature. Putting more carbon into solution than is necessary suppresses martensite start (Ms) and skews the morphology to plate martensite. Plate martensite is not as orderly and regular as lathe and its plates impinge on each other at high irrational angles which can result in micro- fracturing of those plates. This stuff is very high stress and brittle. At 1.00% carbon in solution you will form 100% plate martensite and have an avoidably brittle blade. In solutions below .60% you will have 100% lathe martensite. Between the two extremes you will a mixture, so the idea is to only put the amount of carbon into solution that is necessary to get full hardness and no more.


P.S. I am well aware of shear type transformations, slip systems in martensitic habit planes, transgranular vs. intergranular fracture propagation, etc, etc, etc&#8230; but I am trying to keep this simple. So before any more learned lurkers decide to see if I can keep up with them, I am only concerned with the folks with questions keeping up with this topic.

The grain boundary carbides make the steel very brittle causing intergranular fracture .Here's what the fracture surface looks like .
Attached Images

 

[Kevin R. Cashen]

O.K. now that I have scared you to death with all the problems that can arise, I would like to now explain how you can avoid them and how hypereutectoid steels can be our friend and very useful for the right type of knives.

First of all, it should be readily apparent that if one wants to work with hypereutectoid steel they really need good temperature control for total success. If you do not have the tools or skill for this level of control, a eutectoid like1080 or 1084 is a much better choice.

Now what to do with that pesky extra carbon? Obviously we do not want it in the grain boundaries, or of we have alloying we don’t want it making big lunky carbides within the grains either. To avoid this we just have to watch how we cool it from solution, we need to deprive it of the time needed for it to get into mischief. Slow cooling from high temps give it this time, so when working with hypereutectoid you need to forget all those annealing operations that involve wood ash, vermiculite, or stuffing it in forge for the night. Going any slower than a steady air cooling could give you problems, but if it is enough to beat the curve you can trick that extra carbon into sticking around and forming wider carbide bands in the pearlite or many finer alloy carbides. So for these steels normalizing and spheroidal type annealing are what you should stick with. Once you put that carbon in good places, keep it there by staying below non-magnetic in other heating operations until you are ready to harden.

When you harden you have to keep your temperatures lower than with a eutectoid. Ever wonder why the recommended temperature for 1095 is often given at around 1475F but 1084 can be higher at around 1500F? It is because 1084 can take it, because it doesn’t have that extra carbon to cause problems. The recommend temperatures that bladesmiths love to ignore or ridicule because those irrelevant eggheads in industry came up with them were developed through an awful lot of study on exactly what temperature will put just enough carbon into solution to get maximum hardness and leave the rest in the from of very nicely distributed fine carbides to aid in resisting wear.

So when you have a steel well above .8% carbon be certain to normalize well without overly slow cooling, use sub critical anneals by cycling above 1100F but below non-magnetic (not only will the steel love you, your mills and drills will as well), and be careful to keep the temperatures below Accm in hardening. Unfortunately the best way to accomplish that last one is to have a well calibrated heat source. Lower temperatures and longer soaks are much better for hypereutectoid steels.

Many bladesmiths love to use complex steels and then inflate all the ins and outs into a huge mystery they have some special insight into, like the world won’t truly understand these alloys until they blaze the trail with their silly fumbling. These steels were made to do specific tasks and are understood quite well by rest of the industrialized world. Many of these “mysteries” are simply created by smiths unnecessarily complicating matters by blindly stabbing at answers with inadequate knowledge or tools, with knowledge being the most important tool. Many feel their position as a smith relies upon there being a big mystery around these things, so they scoff at this information as being irrelevant, overly technical or perhaps even robbing bladesmithing of its romance and tradition. But isn’t it nice how easily things can be recognized and dealt with by anybody armed with a few facts.

Kevin,
Could you expound on how forging temps for hypereutectiod steels play into this mix?

J. Scott

Good question. Let's look at the recommended forging temps for two types of hypereutectoid steels, a carbon steel and an alloy, along their lower carbon little sisters for comparison.

1095- 1500F to 2100F
1080- 1500F to 2150F

52100- 1700F to 2100F
5160- 1600F to 2200F

Since there are heat treating operations designed to much more effectively do many of the things bladesmiths like to think forging does, the temperatures really are not all that different on the lower end. Lower carbon contents allow higher temperatures at the upper limits but mostly it is about effectively moving the steel without causing it any problems. We will do all right if we treat forging like normalizing and start out high and work cooler as we go. The more steel you have to move, like heavy profile shaping the hotter you should go to accomplish that. The high heat will break up segregation and other issues from the milling manufacturing process an evenly distribute alloying and carbon. Then as we forge in thinner sections like the tip and edge we go cooler because not only is it not necessary to move so little metal it also is better for the steel to keep things cooler when not heavily deforming the steel.

Regardless of the carbon content, the number one mistake bladesmiths make is forging the entire blade at too low a temperature, whether it is intentional due to common misinformation about low temperature forging, or unintentional from not recognizing those temps. In my classes most of my students will burn some steel on the first day and then get so gun shy of heat that I will spend the rest of the class saying “get it hotter” again and again.

There really are no special heat secrets to forging except to use whatever temperatures it takes to get the job done most efficiently and in the least number of heats. Forging unnecessarily hot will decarburize the piece, over-oxidize the surface and needlessly enlarge grain in an uneven fashion. Forging too cool will not pack edges, but it will increase segregation, create uneven strain, or even result in micro fracturing. Forge you blade in a manner that allows you to get it done in the least amount of heats to result in a blade shaped exactly the way you like with a nice, clean and smooth surface, then move onto normalizing to fix many of the results of forging, continue some of the good things you started and prepare it for the heat treat that will come later on.

 

[Kevin R. Cashen]

You didn't mention about soak time.

When to do it and at which temperature?

Soak times vary depending on your circumstances, but as we often repeat here the good news is that if we have control of the temperature you really don’t have to worry about going too long. It is not wise to take an industry standard like “10 minutes per inch of thickness” and simply divide it by your blade thickness as you may cut yourself short, since the chemistry in your steel doesn’t necessarily understand your mathematics, it tends to follow it’s own. I have found that 5 to 10 minutes is a good minimum, with carbon steels times can be quicker, with alloy steels I would not hesitate to bump your minimum up to 10 minutes or better.

Remember with hypereutectoids it is better to go a little lower and stay there longer. Let me explain why- your soak temperature will determine how much carbon you will put into solution, your time will determine how evenly it will be distributed. If you have an entire grain (incredibly small by our standards, yet a very, very large space by a carbon atoms standards) to fill evenly, you may get the carbon you want in solution really quick with your temperature but now you have to move it though the grains to equalize the concentrations. You can move it quicker by bumping up the heat but this will automatically also put more carbon into solution than you may want.

Think of a single large carbide (carbon chemically bonded to a metal) as an island in an austenite lake. Your heat breaks the bonds of the carbide island so that chunks of its edges are freed up. But those freed bits form a concentrated carbon solution “moat” around the island, now your soak time will allow the freed carbon to spread out through the lake to make it all a solution of even concentration.

You want to quench 1095 in water? Get really good at judging exactly how to low ball the temperature! 1045 or 1050 can be quenched quite safely in water as it will only make lathe martensite of a certain strain or stress level. So if you low ball your heat on 1095, after being careful to take things out of solution in previous thermal treatments, so that only .4%-.5% carbon is into solution you will have basically made it into 1045 or 1050 with unneeded carbide chunks. This is why you always hear smith say that it can be done with a lot of practice and that practice always involves lower soak temperatures, which if you really look at it is a waste of a good hypereutectoid since real 1050 would have been cheaper and easier and those leftover carbide chunks are either wasted or can become a problem. When the same tricks are applied to alloy steels severe under-hardening is the result due to also having to overcome the effects of alloying. It really is best to use the quenchants the steel was designed for if you want to get the most out of it.

I must admit to feeling some frustration when pointing out how certain methods a smith is using may not work only to have them suggest I am overcomplicating things and that it is as simple as heat to non magnetic, plunk it into some oil, and that “works just fine”. It really is so much more complicated than that, and if it were that easy why would we spend all the years we do working it out? If it were really that easy I would have quit long ago out of boredom, instead of realizing that I have a whole lifetime of fascinating challenges to overcome and educate myself about.

It really is that complex, but the great news is that it is also very simple once you arm yourself with some basic knowledge. We don’t need any added mysteries or mystic secrets since the facts, and I dare say “science”, is more even more fascinating and exciting than any of our folklore. It is like my youngest playing with here Tinker Toys (“Connects” actually”) with just two or three pieces she can imagine she has built all kinds of wondrous things, but will bore of it within an hour. But if she adds the whole box of pieces she has weeks of really building wondrous things. It is not more complicated to expand ones horizons.



*******************************************************
I just realized that I got ahead of myself and forgot to mention one of the most important factors in soak time, if not the most important factor- the internal condition of the steel from previous thermal treatments.

Let’s go back to the carbide island in a lake analogy. This is the classic condition you will have in hypereutectoid steel when you buy it from the supplier or mill- spheroidized. With this condition you will have your lakes (grains) of room temperature ferrite littered with hundreds of little carbide islands. This condition will absolutely require a good soak time for the reasons previously mentioned, but the smaller and more numerous the islands, the less time it will take for the lake to be evenly saturated.

If, however you just normalized the steel instead of spheroidizing it, the entire grain will be filled with fine alternating lamellae of iron and carbon which will mix together evenly much quicker. Thus shortening soak time, but also tearing up your tools much worse since wide sheets of carbide are a bit more immovable than tiny beads of carbide. The shortest soak times, if any, will be needed when heating a hardened piece of steel since the carbon is already in solution from being trapped there in order to harden the steel. But this is also not set in stone since much will depend on the rate of heating.

Heating very slowly from below Ac1 (a temp a bit below non-magnetic) to your critical temp will pull carbon out of solution so that you have to soak it again! Heating very quickly will avoid this but will strongly influence the final critical temperature. Since critical temps are determined under equilibrium conditions this means that temps must be constant. Heating the steel very rapidly can actually raise a critical temp as much as 100F due to the lag of diffusion in unequilibrium conditions. Now those presoak temps you often see may be making a bit more sense even beyond distortion in complex shapes.

I don’t bring all this up to confuse or intimidate folks, but instead to drive home the point that approaching these things with too simplistic ideas will only leave you in the dark. Saying that the simplest of methods worked just fine for smiths 500 years ago and thus are fine today, totally ignores the advent of alloying. Ancient smiths had the wonderful luxury of being able to work off the clear cut principles laid out on the most basic iron-carbon equilibrium diagram, just a pinch of chromium completely rewrites that diagram. As with every thing else we live in a MUCH more complicated world than our ancestors.

 

[Kevin R. Cashen]

It's good to understand the iron-carbon equilibrium diagram even though we rarely see equilibrium in the real world. Even though the lines may be moved about due to time and alloying , the principles remain the same.

Those who don't forge have to guess what the original structure of the steel is. Numerous times on this forum someone mentions problems cutting or drilling the steel even though the steel should be 'annealed'. Obviously the steel is not what it should be .The process then is to make it proper by heat treating applying the methods Kevin mentions above.

Originally Posted by bambright
Kevin,

..Between grain refining vs carbon distribution, which one we should do first and why?..

Easy- get the carbon where it needs to be first. Grain refinement is a very quick an easy fix if you know what you are doing.

To complete the picture, I know you have been sharing this before but I would like to ask you again to provide the example of your O-1 heat treatment along with this thread for both with spheroidized and non-spheroidized so we are clear exactly what, how, and why.

Thanks, bambright

http://www.cashenblades.com/info/steel/o1.html

since i'm not equipped for controlled subcritical anneal.. .. i'm curious... because once or twice i've drilled some steel barstock that was suppose to have been annealed... and the drillbit chattered alot... ..which is excactly what i expect from the crucible steel barstock ( i'm vary familiar with that )... but this was from the factory... huh...

Very common, particularly in 1095 these days. A lot of the 10XX series steels I have cross sectioned for under the microscope have a segregated stringer of concentrated carbide running right up the center, often this will include grain boundary cementite so heavy that it can be seen macroscopically. This is why proper normalization is so important and why forging too low to move this stuff around is not improving anything and can even make it worse.

...can the subcritical anneal produce carbides that coarse if the process wasn't watched closely.... what are the limits to the process... or can they run the size of this up to what i'm used to seeing?

What you see in the crucible steel are more the overall effect of carbide networks, very much like the segregated stringers I described in the center of the 10XX series. Spheroidizing can form individual spheroidal carbides that arrange themselves within these networks. If one goes above Ac1 and then cools slow enough to really tap into what is known as the divorced eutectoid reaction you can really gather the carbon together to make very widely spaced and large spheroidal carbides. This will make the softest possible steel for working on but will require an appropriately longer soak time to put back into play.

...
another question... what about tempering carbides.. i know there small but can they be a problem.. .. pending on temp and where they are?
( sorry bout my ignorance but i don't know much about this at the moment )

Tempering carbides don't really come into play until you go to 400F or better. Then enough carbon as gathered together to form a superfine carbide. These initial carbides are often referred to as epsilon and other Greek letters and are not even visible under an optical microscope but their overall effect can be seen as a darkening of the martensite. By this initial nature they are very fine a very widely dispersed so are often much friendlier to our purposes than the previously discussed primary carbides. As you continue to go hotter in tempering the carbides will get courser until they do become visible under the microscope. Then by the time you approach 900F you are begging make large spheroidal cementite particles and we should start calling it spheroidizing instead of tempering. When the fine carbides are not friendly is when they form sheets between the martensite lathes or plates resulting in TME or tempered martensite embrittlment. This is the cause for that annoying dip in impact toughness in tempering from 450F-550F in steels with certain alloying, moly being just one of the culprits.

Subcritical anneal should be explained in some more detail .There are two types.
If we start out with pearlite as would be the case with a low hardenability steel that is normalized, that's case #1. This structure ,sometimes found in the 'annealed' condition is very difficult to machine.If we heat to ~ 1200-1300 F the carbide layers in the pearlite slowly form spheres .The longer the time the larger the spheres .A coarse structure.
The case # 2 is when we normalize a high hardenability steel and get martensite. Heating this will precipitate fine carbides as Kevin has explained.As the time increases the carbides gather more carbon and get larger. This is the mechanism for any precipitation in metals.The earlier stages produce stronger structure as the particles have coherency with the matrix.
Case #1 produces the highest machinability but is not as good for further HT as the spheres require more work to dissolve and diffuse.
Case #2 has finer carbides ,easier to further HT an end up with finer carbides which makes a better blade....So Q&T to anneal ? Sure .

 

[Kevin R. Cashen]

I'm about to forge O1 for the first time today so I'm paying attention very closely. Thanks for all the input and sharing with the community.

Carpenters HT spec for O1
says once at temp for hardening, soak for 5 minutes per inch of thickness and your O1 pages says soak for 10 to 30 minutes. Could you explain the difference in suggested soak times and the reasons?

I based my website numbers on what bladesmith may be doing. Carpenter may assume that you have good controls and have the steel up to temp before counting for soak, I add in a little margin of error for knifemakers in estimating the time at temp and as long as the temp does not go too high it is much safer to err on the loger side than to undersoak. I have found that in salt baths it may take as much as 3 to 4 minutes for the salts to rebound and then I need at least 4 to 5 minutes for the HRC values to maximize. Salts convey the steel temp much more closely and brings the steel up to temp many times faster than air within a kiln could, thus I felt my numbers were the safest route for advising knifemakers.

As a somewhat relevant tip, I picked up a $9.00 multimeter with a temp guage and have it stuck in my forge right now, I'm not sure how high it will read, poor documentation but it is reading over 1000 deg C, so with a really cheap tool one can ahve fairly accurate forge temps for HT.

Good info Kevin.

I have a question though... (purely hypothetical), Lets say you were working with a "hypererectiod" steel with a simple blacksmith type shop set up, and you couldn't nail the subcritical annealing. So,... you slow cooled it from above critical, slowly enough to soften it for stock reduction. What would be the best way to fix it, (if any), before quenching? Could you simply run a series of say three normalizing (air cooling) cycles from progressively lower temperatures with the final cycle being just slightly above non magnetic or something of that nature?

Yes, heat is the ticket. Low temp normalizing may only make it worse since the carbide networks are already there, the carbon will want to move in and join it as soon a the lower critical is reached. However if you go very high, above Accm or the upper critical temperature, you will then dissolve it and put it back into solution, however this will most likely result in grain growth which will then have to be taken care of with a few subsequent lower temperature cycles.

I remember, as I am sure you do, when normalizing was seldom discussed or taught by bladesmiths, this was back when most of the myths surrounding forging, such as edge packing, were the dogma of the day. When I selected Jim Porter to test under I was happy to find that I had picked a master smith who was happy to hear that not only did I know what normalizing was, it was an integral part of my heat treating regimen. This is all worth mentioning due to the irony of ignoring such a critical procedure while assigning virtues to forging that are really only the result of normalizing effects inadvertently occurring during the forging heats, but many undesirable things can happen while forging that only normalizing can properly fix.

O.K. Kevin, thanks. That's what I thought.

However, the "modern edge packing myth" must have had more to do with starting forging from higher temperatures and working through lower temperatures, heating and cooling rates etc,... as the blade became closer to the finished dimensions. This seems only natural in the finishing and straightening heats to avoid or minimize scale and decarb etc. The "old packing myth" probably had more to do with working slag inclusions out and/or work hardening the edge. However, I'm not really sure how all that started.

Excellent info! Thanks again Kevin. Now on to my question.

So, what can be expected when a hypereutectoid steel (let's say 1095) is used in damascus with another high carbon steel (let's say, 15n20)? Let's assume about 66% 1095 and 33% 15n20 (since this works out about right for the common mix of 3/16" 1095 w/ .058" 15n20 from Kelly Cupples...I like to keep things grounded in the practical, as well as what I have in my shop )

What could one expect to see after welding but prior to any other work? The drop in carbon from the mix will bring me close to being a eutectoid steel, but I'm wondering if there are any hidden "gotchas" because I started with a hypereutectoid steel.

Thanks,

-d

You pretty much have it figured out, diffusion will take care of things quite well. Since steel phases will always prefer equilibrium if you have a lower carbon steel next to hypereutctoid material, the carbon will head for that other steel more often than places where it is at. Dynamic recrystallization, where the process is drive along with the help of strain energy of the hammering, will also make it hard for the carbon to settle and keep things in motion. I have a belief, that I intend to prove with testing in the future, that the extremely high temperatures and massive deformation in that making of damascus can have very profound effects upon any segregation leftover from the steels creation at the mill. Of course this is also off set by the addition of other segregation due to the weld zones.

In any case, a good series of normalizations with reducing heats o the billet are always a good idea with damascus, before stopping for the day.

 

[Kevin R. Cashen]

”I have a question though... (purely hypothetical), Lets say you were working with a "hypererectiod" steel with a simple blacksmith type shop set up, and you couldn't nail the subcritical annealing. So,... you slow cooled it from above critical, slowly enough to soften it for stock reduction. What would be the best way to fix it, (if any), before quenching? Could you simply run a series of say three normalizing (air cooling) cycles from progressively lower temperatures with the final cycle being just slightly above non magnetic or something of that nature?”

You may remember. Silly me, I have succeeded to do that for D2 in a temp controlled kiln. Mete and Kevin came with that solution... : http://www.bladeforums.com/forums/sh...d.php?t=602818

I remeber that!

In essence, as far as I know,... there isn't much that it won't fix aside from existing factures, porosity, inclusions, impurities, decarb, scale or contaminations of any kind...

When in doubt,... normalize it with a series of progressively lower temperatures using as few heats as possible, or practical, before quenching it...

The science is interesting, but I guess what I'm trying to get at is, (without getting into a debate over the "idealistic" nature of science versus the "pragmatistic" nature of smithing),... are there some simple rules of thumb in terms of processing the steel, that we can observe regardless of the type of set ups we have... or how to get the most out of what we have rather than a debate over which is better or best,... without being accused of over simplification?…

Couldn't we say something along the lines of,... Avoid slow cooling hypereutectiod steel from above critical except as a last resort, in which case normalize it real good before quenching it?

Much of my point in this thread is that problems can arise by approaching so many variables and possibilities with any type of one size fits all treatments. If we want to narrow it down to very simple approaches we need to then limit our focus to just one issue at a time. If all we want to worry about is annealing then we can say it is as simple as avoiding slow cooling from a high heat. But a whole new set of issues will arise in normalizing, hardening, and forging. The best way to be able to handle all of these situations is to take up the approach of learning the underlying causes and effects in the steel itself so that you can apply that knowledge in any situation. When I am being instructed on something, i.e. “never press that button!”, “why not?”, "you don’t need to know that all you need to do is never press that button!”, but what happens if somebody else presses it by mistake? I really feel it is better fro me to know what will happen and how I can deal with the consequences! Heck I may even be able to rework the system to make the red button less problematic. Thus I am never comfortable with being told "never do this" or always do it just like that" because it only tells me how to follow a recipe exactly. Should anything unexpected happen I am not armed with the information I need to adapt to it. Thus despite the number of times I am often requested to just give a recipe to follow I will try to avoid doing so. If we break the process down and teach how and why it happens we give the tools for people to write their own recipes, that may actually work.

I think that's fine as long as we don't get "lost" in it and lose sight of what we are trying to do, (how we want to do it and why), become more confused or misinterpret the facts and come up with bad recipes or false conclusions. I’m sure you’ve seen that happen as much as I have. It happens...

However, you must have some facts to work with. That's a given!

There needs to be a “pragmatic balance”, between the details and the bigger picture on a more individual and personal basis…

How long a time above Accm to do this?

Mike

Accm is the upper limit, here you are done. Accm is the line between having some carbide left and having none with only saturated austenite remaining, so you wouldn't be to Accm unless you have already obtained this goal, you can look at it as using overwhelming heat to get there regardless of time. If you went there quickly you may keep it there a couple of minutes but if you paced yourself you can consider it done. Remember that since there is nothing but solution there will be nothing to keep the grain boundaries in place and there will be grain growth. Staying there for extended periods will allow the grains to grow to enormous size and require more effort later on to return them to normal.

This is where forging can help. To do this treatment with heat alone and hold it at the temp will result in unhindered grain growth, but if you do it while hammering the strain energy being constantly introduced to the steel from the hammer will drive recrystalization and keep grain growth under control. Proper forging is balancing the rate of deformation with that of the processes occurring within the steel, too little deformation with too much heat and the diffusional processes will get away from you, too much deformation with too little heat and the accumulating effects of strain can get away from you.

I really do want to apologize to any readers who do not have access to a metallurgy text for the use of a term like Accm. I really didn't want to resort to its use but I could not give a specific temperature since it will change with any alloy given. Accm is the designation that applies to this point with any hypereutectoid steel. It is just the temperature where all extra carbon is dissolved, and there are no islands left in the lakes, and no carbide in the grain boundaries.

 

[Kevin R. Cashen]

It still may be preferable to persue the facts. Chances are they can be repeated and confirmed even if there is a risk of misinterpretation.

Can I ask, in general, is it preferable to spheroidize martensite rather than pearlite?

Thanks much for the info and a great thread, Craig

This is truly a great question. Pearlite spheroidizes very differently and much more slowly than martensite, unless it is very fine pearlite. The lamellae of carbide in pearlite will spheroidize like a string of pearls over greater times and the courser the pearlite, the greater the effect will be. So if you made the pearlite by very slow cooling spheroidizing would not be so smooth. Very fine pearlite will produce many more fine spheroids in a shorter time and martensite will produce the finest in the shortest time.

Some may say "well that is fine in the books, but Kevin how can you say you know this with such certainty?":

This is some steel that was pearlitic that I partially spheroidized by heating just through the lower critical, see where the pearlite bands turned to strings of carbide beads in the new solution?

I have many images of spheroidized martensite but none loaded to the net right now, they do not follow lines like this they are instead whole fields of well scattered fine little dots.

He keeps showing the same picture ! But it's a good example.
To clarify - grain growth occurs by a grain absorbing an adjoining grain.Therefore the grain boundary must move.Elements like vanadium collect in the grain boundaries and slow the movement of the grain boundary. Vanadium then doesn't 'refine' the grain. Grain boundaries have larger space for things to happen , good and bad .Such as collecting vanadium or collecting carbides .There also is higher energy in the boundaries so reactions can start there such as the martensite transformation. More grains => more grain boundaries => more initiation points for transformation. Smaller grains gives better toughness.
Another point about carbide size .Comparing 154CM and CPM154 - they have the same chemistry so basically differ only in carbide size. The CPM has smalller and more uniformly distributed carbides. Both makers and users like that. Easier to grind,polish, sharpen. Keener edge and better wear resistance.
That's why the dicussion often comes to smaller carbides and smaller grains !!

http://www-g.eng.cam.ac.uk/mmg/teaching/typd/

Here are a few pages of explaination of the iron carbon diagram and the eutectoid reaction from those nice guys at Cambridge !

“Originally Posted by mete:
He keeps showing the same picture ! But it's a good example...”
That is so true! But to put new images up I need to resize them in photoshop, re-archive them in my home files, run them through the FTP program to cashenblades.com, use up the required space at cashenblades.com, and then get the appropriate link into my posts...
I am just too lazy to go through it all anymore. Also I have many images reserved for inclusion into published material, the stuff I have already put on the net is not as fresh so can be re-used again and again.

Thank you Mark!
Kevin, in one of the recent posts (app 3-4 days ago), you spoke of lath and plate being determined by the amount of heat.
In reference to:
Verhoeven Metallurgy of steel >pg25
0% to 0.6% C martensite is lath.
%C above 1% is plate.
0.6 to to 1% mixture of both.
In getting back to the topic of this thread, of hypo-hyper would you explain if Verhoeven is talking about 1060 vs 1095 steel or if he is talking about temperature and the amount of carbon going into solution.
And would you go into detail on this
1) how heat could be controled to determine a %of c solution?
2) if 1095 could ever be lath? (if it was a 1%c steel instead of .95%) hypotheticaly.
3) a good description of "Banded" steel. (and no, I didn't ask that to make anyone angry.)

Wow there is an awful lot in that short little post. Where to begin...
First my mention of martensite morphologies relation to temperature, the thing is that Ms is a function of chemistry so it is much more practical to look at lathe versus plate as a result of carbon content exactly as you listed from Verhoeven, and that is exactly as you will find in most texts, in which they are simply accounting for the total carbon content of the steel. I have many micrographs of my bloomery iron where you will have grain of pure plate just microns away from grains of pure lathe, and plenty of grains of just ferrite. But in modern steel we really should be able to expect a more homogenous make up and reliability.

% carbon in solution can be determined in many ways, temperature is one, but a real easy way to do it if you have control over the making of the steel is by simply adding alloying that will lock up the carbon. You could easily turn 1095 into 1050 with just a little bit of vanadium. This is the challenge faced by folks who venture into custom melts. "Wow 1095 with some vanadium added would be great! Lets put about .5% in there!", "Uh Ohh, now why doesn't my 1095V harden, even in water??" Because after locking all that carbon up in vanadium carbide you actually have 1045V.

Banding can be the result of a few things, by far the most common is alloy banding. When the steel ingot was poured at the mill it solidified from the outside in with different elements of the alloy solidifying out quicker (study a little on eutectic reactions) and thus causing varied concentrations in crystalline zones within. This is very problematic and the mill will do its best to limit this with time in soaking pits to slow this process, and then get it to the rolling mills quickly to break that crap up and draw it out into less of a problem. But what happens is that this segregation gets drawn out the full length of the final steel into what resembles wood grain. Alloying like V, W, Cr, etc... is substitutional in the way it rests in the atomic matrix of the steel (imagine a stack of golf balls, remove one and jam a baseball in its place) so it does not move around all that easily, this makes stubborn zones within the steel that will be different. When you start moving carbon around with heat treating it will really like to settle in these zones if given a chance, so slow cooling from certain temperatures will make bold bands of carbide an alloying in surrounded by the more ferritic material. This condition has some serious strength drawbacks and industry has developed ways to eliminate it whenever they can. Homogenizing heat treatments will help get rid of it. One can cycle O1 through certain ranges and band it out very boldly and then erase it in one good heat if you know your proper temps.

 

[Kevin R. Cashen]

Wow... lots of info to digest here!

I have been trying to find the answers to these questions on my own, but I'll just throw them in here and see what you folks have to say.

So I take a piece of O1 round stock and forge it to shape, knock the scale off, and throw it in the salt bath at 1700F, then 1600, then 1500, then 1400 (cooling to black between each heat). Rough grind, then heat in salt to 1475F for 15 minutes after the salts equalize. Quench in AAA.

Second piece I forge to shape, run through some descending heats in the forge, knock the scale off, then put in the salt at 1600F and cool to black a couple times. Then heat to 1500F in the Paragon and slow cool (50F per hour). Rough grind, then heat to 1475 in the salt, soak for 15 minutes and then quench in AAA.

Third piece, I take a piece of as received O1 PG barstock, cut and grind to shape, heat to 1475F in the salt for 15 minutes and quench in AAA.


Which one...at least in theory, is the best piece? What did I achieve with each one?

Just for the record, I've done this for realsies... and the hardness all came out the same on my import tester (I think it's the same one you have Kevin). The blades SEEMED to cut the same to me... but there was all kinds of human error involved... I would like to put all three on a CATRA and see what happens.

The first piece... simply cycling in the salts is by far the most efficient in my shop... but I don't know which is best.

Any help and advice is appreciated


THANKS!
__________________
-Nick-

#1 - puts carbides in best condition and minimizes grain size.
#2 - A lot of extra HT to no improvement
#3 - OK if prior HT is OK.

Thanks Mete!

I forgot to say that in #3 the stock was Starrett (if memory serves me) precision ground bar that was supposed to be in a spheroidal annealed state.

Thank you for sticking around here, your career full of knowledge is greatly appreciated!

Nick, you don't really want precision ground. It doesn't do much for you especially if you forge. You do want annealed.You do want decarb free. But precision dimensions cost a lot and doesn't help if you grind or forge it away. McMaster Carr has IIRC a decarb free version at about half the cost of the precision ground.

Nick,
I've noticed that sometimes (with deep hardening steels) just cooling to black may not be enough to cause transformation. You have to make sure it regains magnetism, and for the first couple cycles, this may not happen until it gets down to 500 degrees or so. You probably already know that, but you didn't mention it, so I thought I should.

Oh, haha, I agree mete! I still do quite a bit of stock removal as well, and that was how the third blade for formed. I normally don't do too much forging with O1, but I have a few rounds and do sometimes.

Phillip, good catch! I mis-spoke there! What I mean is let it cool to near room temp as I'm waiting for the salts to get down to my target temperature. I always say black because the color is all gone and they're still toasty, but not too hot to hold in your hand. Glad you mentioned that as "black" just means it's cooled down enough to lose color. Thanks!
__________________
-Nick-

Nick if you are doing the salt normalizations sometime, could you do a favor for me? Clean the salt off some of the steel and give it a heavy etch and let me know if the surface is clean and homogenous. I tried normalizing damascus in salts in the past and found so much decarb and mottling on the surface from the oxidizing of the air cooling surface salts in the etched steel that I have not done it since. But then I finish my blades out to 600X before putting them in the salts. If you ever get an opportunity to give it an etch and find if you have any issues I would then be able to determine if it was just the neutrality of my salts at the time.

Also I am glad Philip was on top of the normalizing with the help of the magnet thing. Deeper hardening steels like O1 have pearlite formation so well suppressed that they need cool more than simper steels in order to get proper recrystallization. This is one case were the magnet is pretty close to fool proof, you wont have ferromagnetism until that austenite is making an new phase.

 

[Kevin R. Cashen]

Everytime I think I am beginning to get it I find out there is another layer under the layer I just learned. Thanks for the great post

That is my source of enthusiasm for the metallurgical study of what we do. After I got my Mastersmith stamp I was concerned about stagnation on a plateau in my career, I feared I was limited to just various combinations of artistic expression and techniques, and knives that would do about as good as it gets with the standard bladesmith recipes I had been working with which left only blind trial an error in their further evolution. That is when I turned to metallurgy and discovered a whole new universe of endless possibilities to explore. I wouldn’t have to simply guess or fill in the blanks with my best assumptions; I had solid answers that endlessly generated new questions and a lifetime of fascinating exploration ahead of me. Just as fact is so often stranger than fiction, the real underlying processes in steel are far more fascinating and wonderful than any of the mythology or folklore I have encountered. And anytime you think you have it all figured out it will set you right on your backside and let you know ho much you still have to learn. I have found that really good information does not make you feel smart, but instead lets you know how ignorant you really are and how much more there always it so learn. Some folks may look to mete, me or others as some sort of light at the end of a long learning tunnel but I am here to tell you that all you are seeing is our flashlights shining back as we look forward into an endless abyss of knowledge yet to be acquired.

The discussion that Nick and Mete had makes me worry...

When using O1, I have been doing exactly as Nick described, making a stock removal blade, soaking it for 15 minutes at 1475 F in my gas forge, then quenching in Parks AAA. I do it this way because, frankly, if I forge it I'm liable to screw the steel up more than I can repair with my current equipment.

Mete's comment that Nick's procedure was "OK if prior HT is OK" is what made me worry. I have been buying Starrett precision ground spheroidized flat stock because I have assumed that the steel company has put their steel in good order before it gets to me. But should I be so trusting?

With the problems that Kevin mentioned with steels coming from the mill with a "segregated stringer of concentrated carbide running right up the center" or "segregation and other issues from the milling manufacturing process", is it really safe to assume that the steel you buy is what it's supposed to be? Perhaps it would be safer to assume the steel is screwed up and to always go through the heat treating procedures that will fix it?

Since you guys use a lot more steel than I do, how much faith would you put in steel from Starrett? Is there a more reliable manufacture that I should be going to?

without going through all the possibilities the most common is improperly annealed material.Drill a hole in it or file it .If it's difficult to do it's not annealed.You have pearlite instead of spheroized carbides.

Now I'm just guessing, but Starret PC 01 is spheroidized annealled. I would think it would be in very good structure as it comes, but I would like to see Kevins thoughts on it. I see Mete already gave his opinion on the matter. For the record, I have never gotten a piece of Starret 01 that wasn't well annealled, but the day may come.

i should point out that this thread is more to deal with modern hypereutectoid steels.... and not to be confused with ancients steels... .. as i think in alot of cases they avoided using ultrahigh carbon steels.. except for crucible steel and that is processed in an entirely different manner to deal with the surplus of carbon..

Mete, in Nick's question of the 3 steel conditions, I have a question?
condition 1 was only normalized with no speriodize
condition 2 was heated to 1500 instead of 1400 missing the lower heat and possibly causing a larger grain than wanted.
condition 3 .. with the same thought that I asked Kevin,
wouldn't you want to hit the temp at 1500 deg since he has the equipment to do so >> to put the most amount of carbon into soultion?
The above are all in question form, and not in challange form. Just using what you and the rest have tried to teach.
Nick thanks for the 3 steel conditions, and in no-way am I useing your part of the post in a bad way!!!!!!

...With the problems that Kevin mentioned with steels coming from the mill with a "segregated stringer of concentrated carbide running right up the center" or "segregation and other issues from the milling manufacturing process", is it really safe to assume that the steel you buy is what it's supposed to be? Perhaps it would be safer to assume the steel is screwed up and to always go through the heat treating procedures that will fix it?...

For the most part much of this is splitting hairs on the final performance in knives. Banding and inconsistencies have always been a part of traditionally milled steels and many knives have been made using many methods which simply cannot result in optimum conditions, and yet the maker and the end users have found them quite wonderful. There are a good number of people using methods that exaggerate and intensify these issues that you have concerns about Chris, and yet their knives are prized by many. Knives really do not push the steel performance envelope as far as all of our marketing and fantasies would have us believe. Most industrial applications really test the properties much more than the sharp piece of steel we carry on our side for cutting twine, skinning a dear or whittling. Compare that to an O1 slitter blade that cuts thousands of feet of abrasive material per hour!

I have seen segregation in O1 but is has been a little more distributed throughout the cross section and only become apparent when heat cycles resulting in banding are utilized. Recently 1095 has been a bit more of a problem however, I have seen some stringers down the center of 1095 that, if filled with excess carbide by improper heat treating, was capable of splitting the bar or blade in two lengthwise in the quench! I have also seen similar behavior in some alloy steel and would not be surprised to see it in 5160. Cold shuts from degassing issues and incomplete fusion in the rolling have resulted in 1084 bars that would split open in the forging. I am certain any number of people reading this could share a story of 1095 splitting, 1084 opening up in working or black stingers cropping up in the polishing of 5160. Every once and a while the makers of simple steels produce some lousy runs, some of it is not fixable and I have had steel replaced when I got a bad bar and could prove it, much of it can be fixed with a few good thermal treatments.

The irony of the situation is that for years we were lead to believe that the forged blade was automatically superior due to the hammering as well as “special treatments” bladesmiths utilized, when the unfortunate truth was that most steel was about as good as we could hope for from the mill and then we proceeded to really mess it up with out forging methods, which makes proper normalizing VERY important in order to get it back to that point. However if we do get a bar of that segregated stuff, all of the low temp cycling and forging that bladesmiths love will only make it worse. Although for so long our greatest anxiety was heating too high in forging, it is in fact the best way to redistribute this stuff.

A big problem we all suffer from is assuming that if a little bit of something is good then a whole lot must be great! In fact every heat has its purpose and none can be used to the exclusion of the other, forging an entire blade below critical is no better than forging one totally above Accm, instead we need to learn when to use each of these heats to our advantage in the overall process.

One very good example of this is the now fading practice of edge packing (relax we don’t need to arm ourselves or prepare for laughter at the mere mention anymore). Edge packing was often described as a short series of light hammer blows just on the edge at a dull red at the end of forging. It was purported to be a grain refining procedure, but it is metallurgical fact that deformation in the 10% range will in fact result in grain coarsening and a rather inhomogenous grain size, while simply losing the hammer for a good normalization will get you there every time. Hence my old slapstick “psychic edge packing” demonstration where I refined grain with heat and my will power alone! I am sure you saw at Ashokan at some time. Despite how silly it was, it still amazes people to see how fine a grain my thoughts could produce in heated steel when compared to the hammer

I guess I just typed an entire page in order to say- relax most steel, particularly O1 an other tool steels are pretty darn good right out of the package, and until you have a problem with the other stuff be confident that proper heat treating will still make a really cool knife:thumbup:

 

[Kevin R. Cashen]

Go back to the reasons for the soak . Starting out with spheroidized structure the soak dissolves the carbides until the matrix is saturated .The carbon then must diffuse throughout the matrix.
Carbides are not all the same .Some dissolve easily [Fe and Cr] but others are difficult to dissolve [Mo,W,V].If the steel has one of the harder to dissolve carbides a longer [and even a higher temperature ] soak will get more carbon in the matrix and better properties.

I am certain any number of people reading this could share a story of 1095 splitting, 1084 opening up in working or black stingers cropping up in the polishing of 5160."
Not to go too far off the track here, but this is the first time I have ever read anything that touches on my experience about 10 years ago with a bar of 1/4 x 1" 1085 that I was forging into a skinner. As I worked the point down , the bar delaminated down the centerline just as pretty as you please. I welded it up ,at a high heat of course, and it turned out well. The high heat required for the forge welding would have accomplished the "homogenizing" of the other wise segregated steel.

Thanks Kevin, that’s a relief. I just hope that all your worrying over the condition of your steel isn’t rubbing off on me.

Since you mentioned Ashokan and said that “Knives really do not push the steel performance envelope as far as all of our marketing and fantasies would have us believe”, I thought of another question to ask. During your Sunday lecture at Ashokan 2008, you showed some metallography of Martensite in steel that had been soaked for a relatively short time, and another of a piece that had been soaked for five hours. The five hour piece had much more Martensite than the other piece. I had always been under the impression that, if we heat treated a knife blade properly, it would end up being mostly Martensite. However, judging by those pictures, that’s not the case. Assuming that I am interpreting those images correctly, what percentage of Martensite should we realistically expect in our blades? Would it be advantageous to try soak my blades longer to make more Martensite (and just learn to deal with the resulting decarb), or, as you mentioned, is that more of the “big problem we all suffer from is assuming that if a little bit of something is good then a whole lot must be great!”??

I was interested in the same thing. Was I misinterpreting those pictures of the O1 soaked for 10 minutes as opposed to the 5 hour soak which showed all that more martensite?

Nothing is 100% but martensite in simpler steels particularly one like O1 is about as close as we can hope. The only parts not martensitic in those images should have been the residual carbides which is not only expected but a good thing as long as they are fine an evenly distributed. In 10XX series steels it is not uncommon to have fine pearlite in the mix and in hypoeutectoid levels there could be left over ferrite (iron). But with hypereutectoid steels with any alloying you should have a solid field of martensite with those fine carbides until the free carbon or alloying gets high enough to cause retained austenite, which definitely is not complete martensite conversion and something to be concerned.

Actually the images of the 10 minute soaked O1 and the 5 hour O1 both contained the same microstructures and each contained essentially the same amount of martensite, the only real difference between the two was the number and size of those carbides. The 5 hour soaked carbide were obviously much smaller and fewer. Now the image of the slightly spheroidized pearlite earlier in this thread is another story. Simply taking an alloy steel to non-magnetic with no soak before quenching will result in very little martensite surrounded by messed up pearlite and ferrite. Which, as we have discussed a previously, will simply dull a new file but still be lousy in the long run.

So we have the two extremes, no soak and hours of soak, the realistic approach is get it up to temperature let it soak at that heat long enough to get things done and quench. That time is realistically on the order of minutes, 5-15, and not hours. This will give you a good even fully hard martensite with a nice dispersion of fine carbides, and it doesn’t get much better than that.

lcf- No worries! If I wasn't willing to get responses (good, bad, or in-between) I wouldn't post in a thread like this. I'll admit it's a tad intimidating putting yourself out there, but I'd rather welcome constructive criticism and learn than keep this all in my shop and possibly never reach optimal heat-treating methods.

Of course I can't find it right now, but I have it written somewhere that the annealing temp for O1 was 1500F. So is it actually 1400?

Kevin, I'd be glad to help out if I can! So how are you normalizing now? I must have missed that.

I went and found my shop notes, and I did the annealing at 1450. So I guess I split the difference between right and wrong

Thanks for clearing that up Kevin, I guess I did misunderstand the slides you showed.

Nick, the Heat Treater's Guide says to heat O1 to 1400 to 1450 F to anneal it. "Use lower temp for small sections, and the higher temp for large sections." Holding time is 1 to 4 hrs, depending on whether its a light or heavy section.

There's more, but I'll just email it to you.

 

[Kevin R. Cashen]

Just in case anyone newer would want to read them, here are the two threads from 2006 about the 5 hour O-1 soak:
http://www.bladeforums.com/forums/showthread.php?t=394824&page=1 Phillip's sample prep and discussion
http://www.bladeforums.com/forums/showthread.php?t=398842 Kevin's sample analysis and discussion
Good stuff.
__________________
mike fitzgerald

...Kevin, I'd be glad to help out if I can! So how are you normalizing now? I must have missed that....

No, you didn't miss it, it is not here. Contrary to what is sometimes suggested I do not feel my way is the only way and am very careful not to put too much of my own personal methods in my posts. I put the proven principles that guide my methods, but for much the same reason I don't make a habit of posting my knives I like to keep my input just about the objective metallurgy and allow those who can use it to develop their own practices without everybody just following yet another recipe. We have more than enough smiths pushing recipes on the way to make the greatest knives, and not enough just giving out the basic information on how the individual can work out their own sound methods, so instead we have a hodge podge of parts of famous guys methods instead of an understanding of why we would do any of it. For example, the guy who after edge packing, torches a blade edge up to temp to quench into goo 3 times before packing in dry ice and acetone, or on the opposite end of the spectrum the guy who goes right for a low temp salt bath in order to make bainite or martemper without even considering why . This is not a critique of your normalizing in any way just an explanation of why a guy like me who shoots his mouth off so often needs to keep his head clear about what he is promoting and what he is not.

I normalise using other heat sources than the salts due to the issues I have had, but I have heard many folks talk about good experience with the salts, making me always want to ask if they could do and etch and let me know if I just had a couple of bad days with the salts, you happened to mention it while I was thinking about it so I finally asked. I seemed to have encountered an oxidizing issue with the salts cooling in air on the blade and then re-entering the bath.

Nick, thanks for not being offended by my using your post. I agree that it takes courage to put it on the line.
In the past I have done the same here also, and then thought the better of it.
1400deg is from pg 533 of the Heat treaters guide. With the range being from 1400 to 1455deg. with a sub note of using the lower temp for smaller parts. Along with a detailed reducing temp step on the cooling.
And from what I read above, Chris is mailing more info to you.
My point was in the "hair splitting" that I think alot of us are zeroing in on is what I was focusing on with the temp questions. Since this thread is past the heat it to red and stick in peanutbutter Q&A.
Then after reading Kevins last post I realized that my question was possibly a little too "to the point". That it would stem more into the recipes answer (for the way I asked the question).
My main goal was to determine if the steel does see a change in that low of temp difference.
The book say's it does

No problem lcf There's not too many things I can't stand back and laugh at myself over.... Wait, I don't think there's anything I can't/don't make fun of myself for
Chris very graciously sent me the pages from the Heat Treater's Guide. I've been wanting to buy it for years, but every time I try to find it I come up empty handed. Chris sent me some links... can anybody tell me if there's an appreciable difference between an early 80's copy or a more recent copy... with knifemaking in mind?

I guess with 1450 I wasn't too far off, but it certainly seems I'd have been better off with 1400.

Kevin- I don't totally understand where you're coming from, but I think I do for the most part. I know you want to help people understand what's going on in the steel and not just give a recipe that someone can blindly follow. And hype has driven some really crazy "legends" in knifemaking.

I was surprised as I thought you used salts almost exclusively for your heat-treating.
__________________
-Nick-

Nick,

This may help looking for Heat Treater's guide. http://www.usedbooksearch.co.uk/ This outfit searches about everywhere or connects to outfits that have huge numbers of sellers.

1982 Ed. ISBN: 0871701413
1995 Ed. ISBN: 0871705206

Don't be drinking anything when you search for the 1995 edition or you'll swallow your tongue, too.

I solved my "want it" problem by getting a copy through inter-library loan and making copies of the material I wanted for my personal reference.

Mike

 

[Kevin R. Cashen]

Kevin in the first part of the thread you have the comment of doing away with the cooling in wood ash, vermiculite, or stuffing in the forge for overnight.
Can we go back to this? AND not for a reciept for spheroidized anneal (well maybe kinda) but an explaination of the two or more methods to get there?
Some who may have read this may be scratching their heads and going what the......
For those with out controlled heat who can't hold just above Ac1 then reduce step the heat. (pg32 HTG.)
For me I have read of heating to the austine region and then quench to martensite and then reheated to just below A1 and went what the......
(pg 32 MoS. Verhoeven) Now I read your (what may be prefered) choice is of the thermal cycling at the sub critical anneal heat levels.
Spheriodize ??? door #1 door#2....
And while your on a roll.. explain divorced eutectoid transformation (DET) I might as well go to work tomorrow with a screaming headache.

J D Verhoeven - Steel Metallurgy for Non-Metallurgists , p35-37
Google 'divorced eutectoid tranformation' and look through the listings for 'books.google.com. You will be able to get online versions.Just check p 35-37.
Keep a copy and read it 100 times and it might be clear.Of course to do it properly there are a number of conditions to be met.

Here's a question I've been thinking about for a while, but haven&#8217;t been able to find someone who's tried it. When using 1095, what would happen if the quenching temperature were set somewhere around 1375 F? There would be less carbon in solution, more carbides, and lower hardness and hardenability. Would the toughness increase (less carbon and lower hardness and hopefully more lathe martensite) or decrease (possible interconnected network of undisolved carbides)? Would the higher carbide volume mean higher wear resistance, or would the lower hardness negate them, like rocks in mud. In general, I guess what I'm asking is why are the lower hardening temps set where they are?

For such a short question one can cover a VERY wide landscape with it. Actually many folks have tried it they just aren&#8217;t fully aware of it, it is well worth noting that there are more than a couple of approaches in blade making that use a few of the concepts involved here, and some historical ones, but they almost always involve hypereutectoid. You see, if you are going to under-oak you need to have something hard in the matrix and carbide is the best you are going to get.

Since you are changing very little with such a low heat what you get out of such a treatment is going to be highly dependent on what you had going in. Those who find the most success with such an approach would be the folks who really put a lot of effort into the pre-treatments. Several heat treating approaches being used today utilize what I call &#8220;carbide games&#8221; where the maker simply lets the chips fall were they may on that overall matrix and allow the carbides to do the work. Such blades will do impressive cuts but in very specialized ways. Your analogy of rocks in mud is a very good one and such a blade would do really well on draw cuts (sawing through rope many times) but would have a very good chance of edge rolling if used for heavy chopping. Imagine a knife made of Playdoh with bits of glass mixed in. It would cut string and paper really well and would even cut you all too well if used just right, but chopping a 2X4 will be a totally different story. But on the string and paper the soft doh would get worn away on each pass to expose more of the glass, making it even more aggressive on such targets and giving the appearance of holding an edge longer, or perhaps even becoming sharper as you use it. However if you then sawed on the 2x4 with the glass highly exposed there will be no support or strength to hold them in and they will simply tear out and the edge will smooth over&#8230; a great cutter but only in a very specialized way!

The opposite of this would be a properly soaked blade that more resembles hard little stones or glass in concrete. The cement will not wear as quickly and thus not heavily expose the stones, so it will not be as impressive on rope and paper but will cut a wide variety of things much longer and be able to whack and chop just about anything you want as well. Now the real trick here is to get those carbides as fine and widely distributed as possible so they do not bunch together and mimic the effect of large chunks. Even in concrete the larger stones get knocked out easier than fine gravel, larger stones also make larger weak boundaries between the stone and the cement.

Whether you get toughness or brittleness due to the condition of the matrix versus the condition of the carbides would rest in your previous treatments. You will not form many sheets on the way up to austenitizing, but will spheroidized carbides as you approach and pass Ac1, these spheroidal particles will then need to approach Accm to then spread back out again.

In carbon solutions of .6% or less you will have almost all lathe martensite and thus greater toughness, but strength is also critical in blade functions (barring bending like taffy in a vice), so you will also have stopped short of maximum hardness and thus the greatest strength potential. It is all a balancing act and ongoing compromise between strength and toughness. For most knife applications if you get the treatments right things work better leaning more toward strength. So many techniques used by makers with simpler equipment lead to &#8220;issues&#8221; that we have been erring in the direction of toughness.

I'm dealing with a hypereutectoid steel. It is W1 with 1.2% carbon (has been tested, also has 0.37%Mn). I've got a reasonable understanding of the processes needed to optimize it from this thread but I don't have specific references for normalizing or austenitizing temperatures for a steel with this amount of carbon.

What I have is this:

1995 Heat Treaters Guide ~ W1 data

Nomalizing ~ 1.10%C to 1.50%C 1600F to 1695F
Hardening ~ Heat slowly to 1400F to 1555F, using the upper end of the temperature range for low carbon contents and lower end of the temperature range for high carbon contents. The range being 0.70%C to 1.50%C

Is it reasonable to extrapolate the normalizing and hardening temperatures from the ranges?

Doing that, I get a normalizing temp. of 1624F and a hardening temperature of 1491F. Seems like the hardening temp. is higher than it needs to be. I understand I will get slightly higher hardenibility with a higher hardening temp. but I don't know if I will get into trouble, too.

What I'm intending to do is normalize 3x's with a quench on #3 and then spheroidize.

Mike

 

[Kevin R. Cashen]

Mike, it sounds like you have a good grasp of what would work well. Higher austenitzing temps result in deeper hardening until you get enough carbn into solution to stabilize the austenite and then the hardness numbers will plummet like a rock, while brittelness is increasing. Normalize at least once at the high end but don't allow things to cool too slowly. That quench on the final cycle before spheroidizing is a good idea (think many very fine carbides)or spheroidizing, do several cycles in the 1300F range. In austenitizing for the quench you then want enough heat to get the desired hardness and no more. Around 1450F to 1475F may be fine.

"The book" would spheroidize 1200F to 1250F, minimum 1 hour. What is happening with "several cycles in the 1300F range" that is different than that? (I'm assuming you mean with air cools between... and so you know, Sarah says, "Several can be more than two and often is more than three, but not many, in any case.".)...

My usual is to stress relieve (1200F-1250F, min. one hour) before heating to quench temp., soaking, and quenching. I feel doing that makes even bigger spheroids and makes them further apart. In this instance, that seems wrong-headed... that I want the softness of spheroidizing but want the spheroides to be as small as possible and not clumped up... to go back into solution, well dispursed, easily. Am I getting it, Kevin?

Mike

I was reading that when the term "water" is used as the quench medium for 1095, what is meant is actually brine.

Does it matter what 'salt' is used? Like can epsom salts be used?

What is everybody's faborite recipe; dissolve salt till an egg will float?

I was kind of taking the "austenizing temperature controls cabide volume and carbon in solution" idea to an extreme. I figure if toughness were needed and 1095, for whatever reason, were the only steel around, higher tempering temperatures would be the best way to go

yes you are getting it, that is why I suggested cycling instread of soaking for extened periods. Quick cycling near Ac1 hopefully will nucleate more spheroids in each clycle and thus make them finer. Starting from fine pearlite or martensite will also result in finer particles but more importantly distribute them very evenly.

Also bear in mind that the finer the spheroids, the quicker they will go into solution for the quench, thus making it easier on yourself as far as soaking.

It is always better to nail it and get maximum hardness in the quench and then temper back to your desired level, not only is it good for the steel it gives you the maker the maximum level of control. I have actually heard otherwise knowledgeable makers ask -why bother hardening beyond 59HRC if that is what you are going to get after tempering anyhow, why not save some time. Such a display of lack of consideration is rather discouraging. The concept is very much like a farmer neglecting his crop to the point it grows no taller than 6" since that is what he is going to cut it to at harvest anyhow .

me2, yes !
69_knives, the most effective brine is a 9 % solution of NaCl in water.The water should be at room temperature. Other salts could be used but NaCl is very available.

God, already back to the middle of page three!!!

The system I'm using is a still-air kiln with computer contol. I've been looking for a reference I've seen, a graph, on thermocouple temperature relation to steel temperature for various heating mediums and can't find it. Thought it was in Moniz' book, then looked in my links... nope. I'm looking because if a person is going to spheroidize at 1300F "several" times they don't need to be having the steel sit at 1300F any length of time. I understand thin sections will heat with the rise in kiln temp. but I'm not sure I don't want to run the kiln up then put the blade in.

Does anyone have references to this information?

Mike