Austempering Advantages?

[itrade]

Recently purchased 52100 from someone who uses Austempering.

As I understand it, the austentizing temp is the same. But, instead of quenching in a cool enough medium to create martensite directly, the steel is quenched and held just above the martensite temperature.

It is held for awhile in an 'isothermic hold'(?) for some special amount of time. This gives the steel a chance to relax before completing the process. Then the steel is brought down to room temperature at which point the martensite is formed.

There are a number of heattreating places that use this method and they talk of all the wonderful advantages in toughness and wearability.

Really 'sounds' great. Do many knife makers use it?

Steve

Several links are on the web. Here is one description:
http://www.ductile.org/dimg/pdf/digest3.pdf

 

[fitzo]

The definiton of austempering in my metallurgy primer defines it as a quench into a medium that holds the temp at 500-600F for an extended period of time, with the intent of producing lower bainite. This can take from minutes to hours, steel dependent.

Marquenching is defined as a quench into the same temp range, held very briefly for equilibration, then a normal quench to produce martensite. This produces an ostensibly tougher martensite, and there is less shock to the piece.

Isothermal quenching is defined as quenching at 300-400F until the martensite percentage is approximately 50%, then holding at 500-600F until the remaining austenite is converted to bainite.

These are the definitions in my book (Meatllurgy Fundamentals, Brandt&Warner, 1999). We have been discussing the various heat treat methods on Paltalk a lot lately. Threads also occur at CKD and the new Primal Fires. The question we have about the "isothermal quench" (actually a misnomer, since it means "same temp", and actually two are involved), is that you are taking the martensite formed and then tempering at 500-600F, essentially, while the bainite forms. We have yet to discover if this will actually produce a tempered martensite edge too soft for edges.

Max Burnette is investigating quenching in 300-400F oil, with an extended hold, and then requiring no temper. His very initial tests on 5160 look very encouraging.

Very intruiging,though, as I think we are venturing into the areas that produce the super-tough blades ala Howard Clark's bainite/martensite L6 sword blades, and probably into those areas which produced the "legendary" blades of Frank Richtig.

 

[rlinger]

UHHHHHHHH, this is GOOD stuff. I'll be following this thread. Don't let no-body let it get away!

Roger

 

[itrade]

Well written fitzo!

That really helps me out.

The process that was explained to me, then, is austempering. The steel was 52100 and the process (roughly) was:

Heat to non-magnetic
Quench and leave in oil that's around 300. (That temp for 52100 prevents martensite from forming, although it seems low to me. Your listed temps of 500-600 sound correct).
Finally allow to cool in room temp to form the martensite

I'm not certain about the temp and time, but you can derive it from the CCT diagram for any steel.

Intuitively this feels like it puts less stress on the steel.

Testing this out will be fun.

Steve

 

[fitzo]

Steve,
That is not exactly austempering, because of the temperatures. It has to be above the martensite start temp to be austempering, as I understand it, which is somewhere between 450 and 400F, steel dependent. If you keep the temp above 500F, it will form lower bainite, another crystal form discovered by Edgar Bain, and will not change with further cooling. It is a much tougher crystal than tempered martensite, but the Rockwell"C" is only about 45-55, and is a bit too soft for the edge holding we want. That is what gives Clark's katanas their incredible shock strength, one of the downfalls of old Japanese swords, which broke far more often than people realize.

I honestly believe that you are getting more of a marquench per my "definition", at that temperature. It isn't exactly marquenching, though, because of the extended hold time, nor is it isothermal quenching by the book definition. It is some very interesting intermediate techique. If you are getting martensite, then it isn't "austempering" according to my book's definition.

Now, the caveat is that I have seen the terms interchanged so often that I'm not sure if my book is correct or just their take on terminology.

The interesting thing here is the hold time, if it is 300F or so. This will form martensite gently. It still gets under the nose of the curve for minimizing pearlite formation. What I have not really been able to accurately determine yet is what happens with extended hold times at that temp. Martensite will continue to form, but how long does it take to reach essentially 100% conversion and what happens after the conversion? Does the continued hold at 300F temper the steel? This is what is being discussed now over on Primal Fires with Max's observations about 5160. It is very interesting, indeed.

Another approach we have been discussing is how to form the bainite backbone as opposed to pearlite and still get a martensitic edge. It seems you can't do isothermal conversion (lower 300F quench til 50% martensite then go to a 550F bath to complete the austenite conversion to bainite) because then the martensite would oversoften with a 600F temper, essentially. Thus, our current thought experiment suggests austempering the entire blade at 500-600F until complete bainite transformation. Clay coat the spine HEAVILY with clay and dry, then torch heat the edge to re-austenize and edge quench to get martensite. Obviously, the typical temper cycle for your steel would be necessary to temper the edge. Our thinking is this would give pretty much the ultimate blade differential HT. However, unless it's a big, rough use chopper or sword, the two-level HT seems unnecessary. A pearlite backbone should be adequate for anything but an extremely hard use, high shock whacking knife.

I am most anxious to hear any report you are willing to give on the steel you have HT'd by the method you have mentioned. Make sure and ask them how long is the hold time, and what any subsequent tempering temp is, if you would, please.

Very cool thread, let's keep it going!!! Metallurgists, pipe up!!!!

 

[mete]

I am a metallurgist, I'm piping up. 52100 -quench 1550F, Ms480F, Mf235F. I am not a knife maker but I really don't understand the benefits of your isothermal treatment. Austempering produces bainite. Once bainite is formed assuming 100% transformation ,no martensite will be formed when you cool to room temperature. The temperatures and times are dependent on the steel. Bainite is tough,in an alloy like L6 as tough as it is with martensite, the bainite is even tougher and for things like swords or machetes or survival knives that is the way to go.Bainite of course will not be quite as hard as martensite.

 

[fitzo]

Mete,
This 52100 thread caught my eye because of some recent happenings with other steels, and I don't mean to hijack this thread, but....

This isothermal treatment is a question we have been beating about lately in several different forums. One maker quenched 5160 from critical in 300 degree oil and held it there for hours. The resulting blade performed remarkably well without a temper cycle. We have been trying since then to figure out what happened. He doesn't have access to metallurgical test equipment, just the old brass rod test, hack stuff, bend the blade, test edge sharpness after heavy cutting, etc. His experiment is so far very limited. The main question here is whether extended hold time at 300F or so would produce a truly superior martensitic structure. The fact it withstood rough use without tempering per se raises very intruiging questions.

I have another question I can't find a ready answer to:
Are TTT curves close enough that you can essentially use the generic curve for a specific steel alloy, or does steel vary enough batch to bacth that the knifemaker has to be concerned? Also, where is a good place to find a compendium of TTT curves?

the other question I have is:
Just what exactly does tempering martensite do to the steel? The reduction in hardness is obvious,as is the increased strength and toughness. What I am trying to find out is, what actually happens metallurgically, as what reading I have done doesn't or can't define the molecular changes.

Welcome to Blade Forums, BTW. We LIKE metallurgists!

Thanks for any reply.

 

[mete]

Fitzo, I do not have the exact Ms and Mf temperatures for 5160 but extrapolating from other info the Mf is above 300F .So he is getting martensite tempered at 300F.That would be the VERY minimum tempering temp,maybe the long time saves him from problems. PLease Fitzo never mention molecules with metals. Metals do not have molecules, or molecular bonds. Metals have metallic bonding, atoms , crystals, and grains. Martensite is a tetragonal crystal( cube elongated in one direction) with an iron atom at each corner and carbon atoms in between. As you temper the carbon forms carbides leaving the martensite. As you temper at higher temperatures more carbon leaves and the long dimension of the crystal shortens until you end up with a cube ( ferrite). I hope you can get the picture from my words. The TTT curves cover the alloy with no change from batch to batch. I couldn't give you a source for the curves and as I am now retired I don't have the info souces I onc e had.

 

[Jason Arnold]

Bainite/Martensite blades... sigh, wish I still had a shop, so I could finally get to selling them once and for all. All those anxious clients... Sure is cool stuff. Damn it.

I have another question I can't find a ready answer to:
Are TTT curves close enough that you can essentially use the generic curve for a specific steel alloy, or does steel vary enough batch to bacth that the knifemaker has to be concerned? Also, where is a good place to find a compendium of TTT curves?

Like mete said, the diagram should be fine. If the batch were to be soooo obscure, then quite simply, you either have a different steel or someone is not paying attention at the mill. Though, cleaner steels (or better said, those closest to the IT diagram's tested steel) are more readily predictable, consistent and parallel to the diagrams results. The respective steel's IT diagram should be even more accurate. This is one reason I request chemical analysis for more complicated work and compare to the given diagram.

Oh and here is a book I use: Atlas of Isothermal Transformation Diagrams for Irons and Steels
It runs $175-200.

 

[fitzo]

Originally posted by mete
PLease Fitzo never mention molecules with metals.

Sorry about that, Mete! Retired organic chemist here, and using my own terminology, certainly inapplicable in terms of metallic interactions! Should have caught that while I was writing. Thanks for the correction, and your reply. I've only recently assembled several books on ferrous metallurgy and am currently learning this very exciting subject in a little more depth. Hope to have some interesting threads with you involved in the future. I spent a year out of college working in research on cemented carbides, so I learned just enough to whet the appetite!

Jason, thank you, too! I might have to try and get the local library to order that book in so I can get a compendium of TTT diagrams for the steels we use. (When I bought a research reprint of Bain's book in at $118, an eyebrow got raised by the "boss". She's a sweetie, but I do need some oversight at times!) It is good to know, though, that the diagrams are essentially applicable to all batches of a given steel. I had seen it written here that "you don't know the specifics of the batch of steel you're working with", and didn't know how much the TTT's would vary, given that, for example, what we call "1084" is actually a carbon range from 0.75 to probably 0.9%.

I left the laboratory behind a year and a half ago, and swore I'd never be interested in chemistry again, but find the allure of understanding the chemistry of what we do with knifemaking very intruiging!

Thanks, again, guys!

 

[Jason Arnold]

About the 1084, what I basically meant was, (using 1084 for an example)... not all 1084 batches are created equal. The range in carbon like you mentioned (.75-.90) is closer to 1080 (.74-.88), that will change things on an IT diagram. AISI 1084 must be within .80-.94 carbon; .60-.90 Mn; .04 P max and .05 S max, to be called such. If it drifts outside those lines, something screwy is up. And yes, that will be effecting an IT diagram and its accuracy for the given steel. Also, dirty steels, usually found moreso in carbon steels will vary quite often in their impurities, sometimes less, sometimes more. Typically this means more or less hardenability. The effect is slight, but just listing an example of how things can change it.

 

[itrade]

Hey Mike you hijacked my thread Just kidding, this is what I was asking.

Here's a great! link to explain some of the stuff these guys are talking about (a great read for all knife makers): http://swordforum.com/metallurgy/ites.html






So, let's see if I got this right:

Austempering is really for creating Bainite. Which isn't a bad thing, if you want toughness over hardness (swords, hackers, etc).

However, for knifemaking tempered Martensite is better.

Then again if you can get good edge holding with bainite then differential tempering or edge quenching isn't so important.

Hmmmm... I think testing some bainite blades is in order.

Steve

 

[Jason Arnold]

So, let's see if I got this right:

Austempering is really for creating Bainite. Which isn't a bad thing, if you want toughness over hardness (swords, hackers, etc).

Yes, austempering is only for creating bainite.

However, for knifemaking tempered Martensite is better.

For edges on knives, definitely yes. As long as you're doing better than HRC56+.

Then again if you can get good edge holding with bainite then differential tempering or edge quenching isn't so important.

Hmmmm... I think testing some bainite blades is in order.

Remember, bainite rarely gets higher than HRC55.

Since I work with bainite composites in sword length blades, my opinion is this... Most KNIVES, even big ones are just fine, fully hardened and tempered that is. But, for something like a sword, differential hardness/treatments are the best way to go. Fully austempered swords are also very ideal. A fully bainitic utility/work knife? No, its pointless and a waste of time. Also, keep in mind, bainite is easier to obtain in some steel than others. So it depends on steel type. Others make it damn difficult in fact. Right tool for the job adage, applies here.

 

[the possum]

Originally posted by Jason Arnold

Remember, bainite rarely gets higher than HRC55.

Since I work with bainite composites in sword length blades, my opinion is this... Most KNIVES, even big ones are just fine, fully hardened and tempered that is. But, for something like a sword, differential hardness/treatments are the best way to go. Fully austempered swords are also very ideal. A fully bainitic utility/work knife? No, its pointless and a waste of time. Also, keep in mind, bainite is easier to obtain in some steel than others. So it depends on steel type. Others make it damn difficult in fact. Right tool for the job adage, applies here.

Thanks for such a wonderful thread, guys! Epsilon- It gladdens me to read your post. I just sent a large Bowie blade to be heat treated, and requested bainite at around 56 Rc. Since it is 24" long and will be used for very heavy chopping and abuse, it sounds like I made the right choice.

It was made from 5160. Is 5160 one of those steels you mentioned that can get above 55 Rc with bainite? Also, just curious, do you heat treat blades for other people?

 

[Jason Arnold]

Possum,

Yes, perfect choice, I have seen a wakizashi that was a HRC56, actually, so full bainite looks promising! For your task, perfect! Wish more guys would get into austempering.

If and when, I get my shop up and running again, I'm considering offering heat treatment services to makers for stuff like austempering or those who may work in the same steels I do. Ie. S5, 1045, S90V.

 

[Galloglas]

Austempering seeks to create Bainite as opposed to creating a Martensitic structure. The steel is brought to critical temp. and then quenched into a media that has been heated to a temperature above the Martensite start point of the steel. For blades this is typically 475f to 600f. The key to getting transformation to Bainite is that we extract heat from the steel fast enough to get behind the nose of the curve yet do not cool it below the point at which Martensite will form. Typically the start point for Martensite is around 450f or below roughly speaking. The steel is then held at this temperature until the metastable Austenite converts to Bainite *instead* of Martensite. This means holding the steel until it converts completely and the closer to 450f you quench, the longer you have to hold it to get complete conversion and the harder the resulting Bainite will be! Bainite made at 500f is roughly the same hardness as Martensite that has been tempered at 500f. If the blade is removed from the quenchant and allowed to cool before it has converted, any remaining Austenite will convert to Martensite as the steel falls below to Ms point at about 450f. This can be a bad thing as the Martensite is untempered this way and will lead to a brittle (but harder) structure. I always temper my blades for 1 hour after they have been removed from the transformation tray (quench tank) and have been allowed to return to room temp. This way, any untransformed Austenite that converts to Martensite as the temperature falls gets tempered.

Bainite is remarkably tougher at any given hardness than tempered Martensite is at the same hardness. Austempering is a much more gentle process and warpage and cracked blades are usually few and far between. Austempering 10XX steels thicker than 1/8” has been difficult for me. I have never tried 52100. The problem with austempering simple steels is that you need to get below about 900f in less than a second to miss the nose of the curve and get the steel to harden. This is tough when quenching into 500f salt. I hear that there are some high tech, high temp. quenching oils that will extract heat fast enough in this temperature range to get the steel below 900f but I have not used it and don’t usually austemper simple steels anymore. With the low alloy steels like 5160, L6, O1 and the like you have a matter of a few seconds to get the heat out and they are simply stunning when austempered.

For instance, I use a lot of 5160 and exclusively austemper….I feel that anything that tempered Martensite can do, Bainite can do better. With the possible exception of edge holding. 5160 that is brought to 1550f and quenched into molten salt (I use NitreBlue from Brownells) at 500f and then held for 3 hours will give a blade of around 54-56 Rc and it will be so tough it will seem like magic. If you quench a little cooler at 475f, you need to hold the steel for 10 hours to get proper conversion to Bainite. At 475f, 10 hour hold we get a blade at about 56-58 Rc and still so tough it is virtually unbreakable.
O1 quenched at 500f and held for 2 hours will yield a blade that is 57-59 Rc and just as tough as the 5160 blade….Austempered O1 is *AwEsOmE* in this hardness range and, in this hardness range, a tempered Martensite blade at the same hardness can literally be chopped to bits with the Bainite blade using impact edge to edge. The Bainite blade will get nicks and shallow cuts in it’s edge. The tempered Martensite blade will have thumbnail sized chunks knocked out of the blade and more than a few full impact strikes in the same location can cut so deeply into the blade that it can be broken.
To get a set in a Bainite blade you would have to put a hell of a big cheater bar on the blade and bend it past 90 degrees. Bainite at 54-56 Rc is actually just a very hard spring. I feel from my testing that blades of Bainite in this range are a magnitude tougher than a tempered Martensite blade at the same hardness. There is *some* loss of edge holding ability in the Bainite blades but this loss in miniscule….I would estimate that the edge holding capability of a Bainite blade at Rc 58 is perhaps 10% or 15% less than a tempered Martensite blade at Rc 58 at the most. What little you lose in terms of edge holding is insignificant over what you gain in terms of toughness and impact resistance. It is a simple matter to stroke the edge out a bit with a stone to get your edge back….a broken or severely bent or chipped blade can be an “end game” situation.

My favorite test for my Bainite blades is to chop concrete blocks till they crumble and strike them against steel Unistrut without damage.

But then again my blades are made to be weapons and tools that need to endure impact, contact, and flexing without permanent damage. Edge holding (for me) is a minor and secondary consideration over a blade that will be used to skin a moose or whittle wood. My opinion is that tempered Martensite is great for skinners and rope and box cutters. For impact knives like fighters, Bowies, and swords Bainite is nothing short of magic.

Try it. The applications of Bainite and austempering for knife makers are many and you may be very surprised at how many of the myths about how Bainite blades don’t hold an edge so well do not hold up under extreme testing and impact tests.

Brian

 

[mete]

Brian, thanks for the numbers and description.I've always told people that there are two problems with steel -1-picking the steel, 2-heat treating the steel.

 

[itrade]

This is great information guys, just perfect thanks.

So much stuff to try - so little time.

For sure I'm going to try some bainite blades.

Steve

 

[rlinger]

Hey Fitzo,

Where should I look to get that book. Is it going to cost as much as a new machine??

Roger

 

[fitzo]

Hi, Roger,

The metallurgy primer I bought was through Amazon.com used books associates and I think I paid about $15 for it in like-new used condition. The specifics are "Metallurgy Fundamentals" by Daniel Brandt and JC Warner, c. 1999 by the Goodheart-Willcox Company.

The coolest of books I've found so far is Bain's. I got it through an Amazon associate, too. I paid dearly for it, and waited 8 weeks while it was reprinted from a library copy. Since this is a one-off type thing, the price was exorbitant at $118. Not terrible for a technical book, of course, but I had to clear it with the boss.
The full title is "Alloying Elements in Steel" 2nd Edition, Edgar C. Bain & Harold Paxton, 1966, American Society for Metals. This copy was reprinted by UMI Books on Demand. It is a softbound book; the reprint is all very clear and the graphs entirely legible.
It was recommended to me by Howard Clark as the "one" book to have if one wants to understand the crystalline transformations of steel and what effect differing alloys have on the steel. Great book, highly technical, definitely a choice I am glad to have made. A friend found a copy by accident in a used book store for 50 cents!!! Naturally

One other really cool book is "The Making, Shaping, and Treating of Steel" by US Steel, c 1957. I paid about $30 for it. While this is more comprehensive about the whole process of making steel and how it's worked, it does have some good metallurgy information. What's sort of cool is all the 50's photos of the steel process. Very dated.

Hope this info helps. I'm a book nerd, so no matter what I get involved in I always assemble as much print info as I can afford on the topic.