Improving Heat Treat Knowledge

F

FunkCoaster

Joined
Nov 2, 2010
Messages
172
I'm a bit of a data nerd, trying to categorize steel heat treat (HT) operations. The goal is to present a structured, easy to read recipe of "Here is how to harden 1095", "Here is how to anneal A2". These questions are fairly repetitive on the forums, so this effort tries to address that. I will also use these recipes to track what I have done to my own blades.

If this seems plausible, I'll turn it into a free website for bladesmiths to use.

HT Operations

These are procedural, so a "heat" can be a kiln preheat or the heating of steel, depending on when it happens.
Each operation can include notes that describe special instructions.
The operations can repeat, so multi-quench, interrupted quench, and other operations can be accounted for.

- Decarb protection: Using a foil wrap, salt bath, etc
- Heat : Rate ( degrees / minute ) and target (degrees)
- Blade put into kiln
- Quench : Medium (water, brine, plate, air, oil)
- Wait : Duration to wait after the previous operation.


Issues and assumptions

- A kiln is assumed. I assume that doesn't generally impact the operations used for torch, forge, salt bath, etc. I realize that solid fuel forges, fuel rich/poor propane forges, and other factors may impact a recipe, but I will ignore them for simplicity.
- Will a given recipe always be correct? No! Variations in alloying, steel batches, and other variables mean that the same recipe may not always work the same.
- Can there be multiple recipes for the same steel? Yes! For various reasons an air quench might be used instead of plate quench. That makes the recipes different, but not invalid.
- Every operation can link to a forum, wiki page, video or other resource to provide further information.

Feedback I'm looking for

Are any operations missing?
Is this a fool's errand?
Do you see any value in this?
What would make this effort really helpful to you?

I'm looking to give back to this community in the ways I can. I'm way less experienced/educated than Stacy Apelt, can't HT like Kevin Cashen, and have fewer dogs than Nick Wheeler ;).
But maybe I can help some noobs by making a dent in the HT data arena.
 
I think it would be useful. Do one or two using the information you have available and let's see how it looks.

- Paul Meske
 
not sure if you want input or do it all yourself. here is mine for O1 and 1.2519

O1
Pre-heat furnace to 1475F
Have quench oil at 120F (or what oil maker says)
Add blade and stabilize (furnace temp back to 1475F. Blade and thermocouple same color.)
Soak 8 minutes for 3/64" or thinner 10 minutes for 1/16" to 3/32" 15 min for 7/64" to 9/64" 20 min for 5/32"(thickest stock i have HT'd)
Remove blade and quench in oil to cool enough to touch.
Optional for 1/16" and thinner: quench in oil to black, then weighed plate quench to room temperature
Wipe blade clean.
Optional put blade in freezer for 10 minutes
Have tempering furnace at required temp. 300F: Rc63-65 350F:Rc62-64 400F:Rc61-63
Note: O1 can have hydrogen embrittlement when tempered between 400F and 550F, so if you want a blade softer than Rc61, find a different steel
Temper for at least one hour.
Cold water quench
Temper for at least one hour in over 25F cooler than first temper
Cold water quench

hope this helps.
scott
 
Last edited:
Scott,
For recipes, I definitely will not be doing it all myself. I'll try to put your recipe in a format and see how it goes.
 
While I believe it will appear to simplify things for knifemakers, I can't help but think there are consequences to spoon feeding recipes to folks. Why not instead, figure out a easier system to introduce people to the basic metallurgical principles that will allow them to recognize, formulate and troubleshoot the heat treatment process. There are already so many HT recipes to choose from... what we need is a filter, not a feedbag.
 
Rick Marchand said:
While I believe it will appear to simplify things for knifemakers, I can't help but think there are consequences to spoon feeding recipes to folks. Why not instead, figure out a easier system to introduce people to the basic metallurgical principles that will allow them to recognize, formulate and troubleshoot the heat treatment process. There are already so many HT recipes to choose from... what we need is a filter, not a feedbag.
Click to expand...

a basic recipe was a big help when I was a nubie. I have heat treated about 100 O1 blades since then and the only change to the process has been soak time for thin stock. Compare what I wrote to what Crucible, Carpenter, Metal Ravne, Bestar, Precision Marshall, and Cincinnati Tool Steel Company has on their websites and again the only difference is soak time. I have also found that the same recipe I posted has worked well for me with 1084, 1.2519, Simonds O2, and 80CrV2.
I know some folks love triple quenches or 8 step normalization or ........, but a list of basic recipes is a good place for a newbie to start.
scott
 
Scott, I get what you are saying. There many websites that can give new folks decent HT regimes for most steels... but there is always play and things left out. I did not see any mentions of pre-ht thermal cycling in your recipe. Perhaps you get your steel from a reputable supplier who guaranties uniform grain structure. Most simply offer barstock in a highly spheroidized state. I don't see that as optimal. You make no reference to the presoak industry standard recommends. You mention a freezer(prior to 1st temper?), dropping the second temper by 25F and cold water quenches(after temper cycles) with no info as to why. You obviously tweaked a few things to get the results you wanted and that's cool... but we are left to blindly follow your recipe.

I use a different regime for O1... are they BOTH going to be put into the database?

Please note that I only pick apart your recipe to point out that there are several ways to go about it and not everybody agrees on the best way. But we can all agree that basic metallurgical principle will remain constant.
 
Last edited:
Rick Marchand said:
While I believe it will appear to simplify things for knifemakers, I can't help but think there are consequences to spoon feeding recipes to folks. Why not instead, figure out a easier system to introduce people to the basic metallurgical principles that will allow them to recognize, formulate and troubleshoot the heat treatment process. There are already so many HT recipes to choose from... what we need is a filter, not a feedbag.
Click to expand...

I like Rick's idea. There are plenty of places out there that have simple heat treat instructions for the novice that are specific to knifemaking. I know AKS has one and Kevin Cashen has some on his website and there are other suppliers with the info as well.

To Rick's point there isn't a whole lot out there on digesting the dense and dry metallurgical information available. Verhooven's book is about the closest I have found and it will definitely make your brain hurt if your not familiar with it on an elementary level.

I'd like to see something like a weekly topic on metallurgy. Like "Carbides what do they do and how do they interact with your steel", next week "Grain size; what does it mean and how can you alter it", etc. And keep them all in a master sticky thread.

I don't know, half asleep here, but I think Rick is on the right track.

-Clint
 
Rick Marchand said:
Scott, I get what you are saying. There many websites that can give new folks decent HT regimes for most steels... but there is always play and things left out. I did not see any mentions of pre-ht thermal cycling in your recipe. Perhaps you get your steel from a reputable supplier who guaranties uniform grain structure. Most simply offer barstock in a highly spheroidized state. I don't see that as optimal. You make no reference to the presoak industry standard recommends. You mention a freezer(prior to 1st temper?), dropping the second temper by 25F and cold water quenches(after temper cycles) with no info as to why. You obviously tweaked a few things to get the results you wanted and that's cool... but we are left to blindly follow your recipe.
I use a different regime for O1... are they BOTH going to be put into the database?
Please note that I only pick apart your recipe to point out that there are several ways to go about it and not everybody agrees on the best way. But we can all agree that basic metallurgical principle will remain constant.
Click to expand...

I understand and am not taking it personal. As far as steel, all the O1 is use is name brand precision ground flat stock, mostly Starrett. as far as skipping pre-soak and cold water quench between tempers, that advise came from Roman Landes on another forum. the freezer part was my idea to get the steel to Mf as quick as possible and why I put optional next to it. i use mostly thin stock, 3/32" to 1/32". I have tested numerous blades after a one hour temper at 275F and ended up with average hardness of Rc64-65. would be nice to see your recipe. I know where there are some lists of recipes, the easiest to follow and includes science to go along with it is http://www.cashenblades.com/heattreatment.html.
scott
 
Last edited:
I like this idea that you are working. I want to make a suggestion. For your quench section, I think that will be very tricky as you would need to point out different methods for achieving different goals (regular quench, marquench, austemper, etc.).

It might be easier to:

1) Explain what the quenching process does
2) What cooling rate is required (use a few points from TTT: Pearlite Nose, 400 C, Martensite Start, etc)
3) How to determine the cooling rate (heat transfer equation + an example)

Then you could add a few example for those who just need a basic quench procedure without delving into the science.

As an example, I've been working on a sword that I have yet to harden. I did some math and found out that based on 1/10" thickness for a flat bar of O1, I need a heat transfer coefficient with a minimum of 500 W/m^2 K (watts per square meter per kelvin) to be able to cool below 1100 F within 7 seconds from 1550 F. There is no real maximum but if the heat transfer coefficient soars above 3500, the temperature difference between the core and surface are no longer insignificant. Therefore a value in between these two extremes would be ideal to avoid the pearlite nose and to minimize uneven contractions/expansions.
 
Is anyone allowed to chime in and give their thoughts? Maybe best to have a metallurgist regulate this input? Just some thoughts....I see this as a big endeavor. Lots of info to compile. The annealing, stress relieving, hardening, tempering, cycling, quenching, forging, spheroidized structures, all for every alloy we commonly use. Quite a bit of data that needs some amount of moderation by folks who genuinely "get it".

Like the pre-heat industry standard recommendation (ie: heat to 1200, hold for 15 minutes, heat to 1475, quench). Differing opinions, depending on alloy. Then it doesn't become a matter of opinion with certain alloys...it's recommended (highly). O1 for instance. Industry "standard" may recommend a 1200F pre heat and hold. But to my knowledge, O1 steel with knife size cross sections does not need the pre heat, and for knife cross sections it is best to skip it and do full ramp to target temp, let the oven settle into that temp, then put blade in. The pre heat on O1 is for items to be quenched that have odd cross sections. However, move over to A2, and a lot of folks do recommend the pre heat and hold before ramp to aus temp. Personally, the reasons I have seen given for pre heat of any alloy is dimensional stability in complex shapes, or very thick cross sections, because it takes more time for the entire piece to equalize. Maybe the higher the temp, and the more alloying, the pre heat is more necessary? I don't know. But generally speaking, knife cross sections, low alloy steels, it is best to have the oven AT aus temp, then insert blade, so that it heats as fast as possible.

As far as industry standard HT and knives we are trying to make, there is a disconnect. For example, 52100 (a ball bearing steel) is often recommended to be hardened at 1550F. This will make a good ball bearing, but not so much a great knife HT. More carbides are dissolved, retained austenite is higher, an overall lower HRC reading, possible larger aus grain. I've seen industry standard HT of 80CrV2 from 1575-1625. WAY too hot for a knife. But if all you're after is wear resistance, at the expense of toughness, maybe you want that HT?

The whole spheroidized topic is hard to nail down. It would be cool to have a chart with steel and supplier. Coarse spheroidized vs fine spheroidized. I am under the impression that when you buy O1 from online vendors such as flatground, speedymetals, etc, O1 will always come fine spheroidized, and does not need any normalizing or thermal cycling procedure. Ready to machine, ready to harden as is. You can if you want. However, buy 52100 from a particular vendor, say NJSB, and you'll need to normalize and then subsequently thermal cycle, due to it's coarse spheroidized matrix. However, buy 52100 from AKS, I have been told by Chuck it is ready to machine, ready to harden, as it is fine spheroidized and sourced different than NJSB. I was under the impression that NJSB 1084, 1095, W2, 52100 were all coarse spheroidized, but I don't think that is the case. It was just the 52100. So I got into the habit of normalizing and cycling any and all of the steels I received from him, just in case. It would be cool to verify and say exactly what steel from what vendor is in what condition as received. There is a bit of time and energy involved in cycling steel, it would be nice to not do it if it doesn't need it.

How does one tell if a steel is fine sheroidized vs coarse spheroidized without the SEM? (and besides trying to harden it as per usual with lower than expected RC numbers) I have no idea.

About the freezer thing, I do it as well. It probably doesn't do anything beneficial, but we understand Ms-Mf. So if you don't have dry ice solution or LN, and you're working with low alloy carbon, even tho you keep aus temp at 1475 (which should put just enough carbon in to reach max RC with little to no RA), then why put it in the freezer if RA is negligible? Obsessive Compulsive Disorder? Maybe, I do it just in case there is some RA. Is not 0°F better than room temp? I think so, just like I think dry ice is better than a home freezer, and LN is better than dry ice. Is it needed for low alloy carbon and tool steels? The metallurgists say NO (ceteris paribus...proper aus temp etc). I do it anyway. Just like I normalize and cycle AKS 52100.

Heat transfer coefficients, I think is probably too much science than what is needed for something like this page we are talking about assembling. Understanding pearlite nose and quench speeds (nickel ball), these are more helpful, I would think, and is probably sufficient for our needs without calculating an exact cooling rate for a given oil/steel/temp etc. However, to eat my own words, those calculations may be cool to understand if you're approaching mass that is pushing the cooling rate of the oil. Like P50 through hardening a 3/8" piece of W2. Will it? The calculations would be helpful there, I would think. Generally speaking, tho, matching alloy to "best" quench medium would be cool. But here we get into some sort of uncharted territory. We all know A2 is air or plate quenched, right? Not so fast, there may be some benefit to quenching it faster, as in a medium speed oil, minimizing RA and certain unwanted precipitates. But I think generally speaking, the higher the alloy, the less need for a fast quench, and going faster is only going to cause problems more than help anything. All of that contingent upon grain size, even. The finer the aus grain, the lower the hardenability, the faster the quench needs to be.

So much to consider, and I'm not the one to be providing the metallurgical info. I am not qualified, but I can tell you what I do and why. It may not be standard, or even the best, but I do strive to learn to do the best.

I have been told by metallurgists that fine pearlite is the best condition to harden from. Fine spheroidized is OK, but needs a short soak to put that carbon back into solution.

For recipes and such, I can show you what is done in my shop. O1/O7 come fine spheroidized and are ready to machine/harden. Coat blade with ATP-641. Ramp empty kiln full throttle to 1475°F. Kiln equalizes, insert blade. Once read-out has rebounded to 1475, begin soak of 15 minutes (.040"-.125") 20 minutes (.125"+). Quench in medium speed oil for O1, fast oil for O7 (1.2442), agitate, remove before Ms and straighten as needed, back into oil until ambient. Place in freezer for 20 minutes, then temper at least twice, one hour each. 375F for 63-64. 400F for 62-63. 440F for 61.

52100 from NJSB (all cycles with ATP-641), normalize: equalize kiln 1700°F, insert blade, once readout had rebound, start soak time of 20 minutes, air cool. Normalizing complete. Thermal cycle: 1600°F 15 minutes, air cool. 1475°F 15 minutes, air cool. 1475°F again 15 minutes, air cool. 1425°F 15 minutes, air cool. Thermal cycle complete. Harden: 1475°F 15 minutes, quench in P50 (I use Parks for 52100 instead of medium speed oil..no issues...others have experienced issues with micro cracks on surface), agitate, straighten before Ms as needed, back into oil until ambient. As quenched should be 67HRC+. Into freezer for 20 minutes, then temper at least twice, 1 hour each. 375F for 64, 400F for 63, 425F for 62, 450F for 60-61. Please note my HRC numbers are educated guesses. Pretty close, tho.

To spheroidize anneal: soak at 1475°F 15 minutes, quench. "temper" at 1200°F for 2 hours (1 may be enough). This is a "bladesmith" spheroidizing, and will produce a fine spheroidized structure that, in all reality, to me is just as soft as coarse spheroidized steels.

Hope this helps a bit, just thought I'd give my 3 pesos worth. BF has done a LOT for me in the past.
 
Last edited:
The best place to start is with the specific alloy data sheet(s). Uddeholm, Carpenter, Niagara, Crucible, Latrobe, and others provide these and there is a lot of usefull information in them.

John Verhoeven's book Steel metallurgy for the non- metallurgist is a great resource. Also recommended is Tool Steels 4th and 5th edition.

Steel varies from batch to batch and from mill to mill. It is unwise to have one recommended recipe for each type of steel because of variances in furnaces and chemical composition of steel. The size of the knife will also have an impact on how long soak times should be.

To become a successful heat treater requires a lot of practice and lots of test coupons.

Heating rate, cooling rate, prior steel condition, proper quench medium, sub zero/cryogenic quench, single, double, or triple temper, upper temper, lower temper, pre- quench, multiple quench, single quench, normalize, anneal, etc., there are many factors that require practice and research.

Hoss
 
Last edited:
scott.livesey said:
I understand and am not taking it personal. As far as steel, all the O1 is use is name brand precision ground flat stock, mostly Starrett. as far as skipping pre-soak and cold water quench between tempers, that advise came from Roman Landes on another forum. the freezer part was my idea to get the steel to Mf as quick as possible and why I put optional next to it. i use mostly thin stock, 3/32" to 1/32". I have tested numerous blades after a one hour temper at 275F and ended up with average hardness of Rc64-65. would be nice to see your recipe. I know where there are some lists of recipes, the easiest to follow and includes science to go along with it is http://www.cashenblades.com/heattreatment.html.
scott
Click to expand...

I didn't intend to imply that your O1 heat treatment was wrong, just that trying to narrow it down to a few recipes is difficult. In its simplest form...

O1
- Austenitize at 1450F-1500F, hold for 10- 30 minutes
- Quench in warm, medium speed oil.
- Temper between 400F-450F.

Anything beyond that requires a more advanced understanding of basic metallurgy or a more detailed explanation as to why you are doing it. Otherwise, you are blindly following a recipe. Perhaps, Roman explained it to you... I don't doubt that. But everyone else has to take your word for it. I would just like to get folks to the point where they can reason out... "O1 is a heavily alloyed steel with an excess of carbon. It makes sense that we soak at heat to allow those extra carbides to get into solution."

If HT beginners were encouraged to acquire a basic understanding of things like the Iron Carbon Equilibrium Diagram, How alloying affects steel, Grain refinement, what TTT charts are and how to read them, etc... there would be less bogus info spread about the forums and hammer-ins. Everyone wants the quick answer. Well, if you are lucky enough to stumble upon it, good for you.... you are still none the wiser, have no way to troubleshoot your process and no way to explain how it works to the next maker. Not far off the "telephone game" you played in elementary school.

Maybe I'm just trying to complicate a non-vital part of knifemaking.... hmmm?
 
Some responses, in no particular order:

" It is unwise to have one recommended recipe " - absolutely agree. Notes and text would be helpful to qualify recipes. These might say what blade dimensions, what source for steel, what known state (ie spheroidized), or other considerations.

I've read Verhoeven's book, two general metallurgy books, and read the forums for 8 years and can now confidently say that I don't know much. But it's evident that even seasoned makers use different recipes, and this idea may provide a consistent "thing" to point at while criticizing, improving, or sharing. BladeForums is great, but it can sometimes be difficult to find what you're looking for.

Will this spoon-feed new makers and prevent them from gaining the depth of knowledge to make great knives? That's up to the maker. Aaron Gough provided a literal shopping list for the items needed to get started along with a how-to. I used that as a launching point. Each recipe can also link to relevant topics, reference material, and other further reading to help makers continue their education. And I think that linking to topics like "What the devil is a carbide" will help.
 
You must log in or register to reply here.
Back
Top