View Full Version : What is gained by Triple quench?

DiamondG Knives
08-22-2004, 03:44 PM
Ive been doing some reading, and find refrences to triple quenching. I understand that with 52100 muli quench is the norm. My question is will 5160, 1095 and 1084 benifit from doing multiple edge quenches? Also are multiple tempering cycles needed?? And while Im at it, will these steels benifit from Cryo treatment? And if so, at what stage do you do it?

Thanks And God Bless

08-22-2004, 05:19 PM
Mike, triple quenching is used to reduce the grain size of the austenite from which the martensite forms, thus improving toughness. It also dissolves more of the carbides, putting more carbon into solution, thus improving hardness. You must use progressively lower tempertures to austenitize so it takes some degree of instrumented control to really do it right. If your target austenitizing temperature is 1500F, quench first from 1600F, then from 1550F, then from 1500F. However, you can eyeball it and do some good. Tempering between quenches is not really necessary but re-heat gently to prevent thermal shock.

Cryo treatment is used to transform any retained austenite to martensite so it is a continuation of your quench. High carbon (.95% and over) responds best to cryo treatment as the retained austenite is caused by high carbon content. 5160 would probably not show any improvement. :101

08-24-2004, 10:22 PM
So a cryo treatment could also be benificial for 1095? I was under the impression that it was meant for stainless steels. Or at least all the talk I've ever heard about it was on the topic of 440c or 154-cm.

Ed Caffrey
08-24-2004, 10:39 PM
As quench said, cryo is just another contiuation of the hardening process. Any time you can convert retained austinite to martensite, you doing a good thing. The average knife user often would not be able to tell the increased performance with simple carbon steels, but when there is a small amount of chormium present, as with 5160 or 52100 the results could almost be called dramatic by comparison to a blade of the same material that has not been cryo-treated. Generally I can achieve about 10-15% increase in cutting ability with 5160, and as much as 20% with 52100 blades that are cryo-treated versus blades of the same material(s) that were not cryo-treated. Pretty significant when you think about it.

08-24-2004, 10:56 PM
Sorry for having to ask so many questions on this topic but I feel it'll aid me quite a bit. So here it goes. Well talk about 5160 in this post, keeping in mind the other steels mentioned, but I'm mostly concerned with 5160 right now, as i have alot of it on hand.

Your saying a cryo would increase edge holding ability? Would this at all affect over all toughness?

I'm mostly concerned about this because my favorite knives are large camp knives and bowies. And this 5160 is great for these knives as far as toughness and shock resistance goes. But if theres a way to increase edge holding also, that's always a plus.

Now I'd like to know if its possible to do the cryo quench without being setup with liquid Nitrogen and anything like that. I do most of my forging and quench type work outside under my carport, and also have a small (12x25) shop that i do my finish work in. Is it possible to do the cryo in these work areas??

Thanx for all the help,

08-24-2004, 11:00 PM
Now I'd like to know if its possible to do the cryo quench without being setup with liquid Nitrogen and anything like that.

You can cryo with acetone and dry ice, but that's probably more expensive in the long run then LN.

Ed Caffrey
08-25-2004, 09:15 AM
My short answer to doing cryo without liquid nitrogen is no. Dry Ice and Acetone can be a hazardous mixture to use. The other drawback is that it will only achieve about -70F, which would take an extraorinary amount of time to achieve any conversion. There are several schools of thought concerning this, and this is just mine, but I have experiemented with the process and found little benifit from trying to cryo with dry ice/acetone.

Cryo treatment will often times cause an Rc hardeness test to read a couple of points higher but you have to remember that your converting austinite to martensite. If you worried about the extra hardness, my suggestion is to temper the blade after the cryo treatment at 25F higher than the previous temper cycle(s).

K Juedes
08-26-2004, 09:58 PM
Wouldn't a cyro treatment warp or crack the blade because it would cool it too fast?

Maybe i'm just confused. :confused:


08-27-2004, 12:38 AM
The book (figurative) says yes it can cause micro cracking and advises when doing with liquid nitrogen to introduce it slowly to the steel (some form of heat exchange that can vary from simply entering drops of LN into the cavity to more elaborate methods). I have read posts of real experience where bringing the steel very slowly down to cryogenic temperature is more beneficial. We ol'knife makers normally can not afford such equipment expense. I certainly can not. I have toiled over two schools of thought on how to hand lower a blade into a LN dewar. Some think it best to go very slowly. I prefer to put her in there as quickly as possible so that all the surface gets the shock at pretty much the same time. To my knowing, I have yet had one crack or warp. I think it much more likey to crack than warp and the cracking may be so small as not to notice before the knife is delivered.

It is important to remember that cryogenic treatment is not a seperate procedure outside the heat treat but instead is a component of it. A heat treat can be done without it but if it is incorporated it is part of it and not something that is done in addition to it.

Complete martensite obsorbtion (not the correct term, I am sure) is about 0.80 percent carbon. I believe any steel containing enough carbon content can benefit from cryogenic treatment regardless of how well the initial part of heat treat went, up to and including quench. I say this because it is so difficult to achieve a perfect heat treat. As stated above, some lower carbon steels, such as 5160 as an example, most probably will not benefit at all. I have tried it with such steels with no noticable difference. Even though they are specified as high carbon they are low by comparison.

Triple quenching appears to refine grain and therefore popularly refered to as giving credit to the quench. Actually, the grain is refined during soak - not the quench. I am still at school here reading you guy's stuff about triple quench however, I think multiple soaks followed by quench may better refine grain if each and every set are done correctly. In other words: I'll stick with a proper single soak but will not close any doors behind me concerning this.


08-27-2004, 08:15 AM
Correct me if I am wrong but dont the professional Cryo places have a system where the part is actually never in contact with the LN. I seem to remember reading in The Wonder of Knifemaking that there were some less than desireable effects from the blade surface coming in contact with the liquid. Has this theory been debunked?

08-28-2004, 06:56 AM
Joe, I think you are correct. The folks at 300Below let the vapor from the LN do the chilling and never actually immerse the part into the liquid. The more gentle cooling and heating are said to minimize distortion. Remember that since the part is now below the Martensite Start temperature, the speed at which it is cooled is no longer critical. Only the temperature to which it is cooled affects the amount of martensite formed.

Now I have a question: has anyone ever heard of a maker using multiple austenitizing, quenching and cryo treatments? Multiple normalizing and multiple hardening treatments are common but I wonder if there is anything to be gained from cryo treating after each heating and quenching cycle. :confused:

08-28-2004, 09:50 PM

There is/was a knife maker in Rochester, Wa that does (or at least did) cryo after each heating step of the treating process. The last time I talked to him was about 4-5 years ago and he was claiming that he cryo'd after the quench and then after each of three tempers (useing ATS 34) and felt he got 2-3 points higher Rc and a much tougher blade that held it's edge far better than any other process he had used at that time.

Don't know if he still does things this way or if he really got the claimed benefits. Might be wroth trying if I can ever get the time.


08-28-2004, 09:53 PM
Whoops-------meant to say that he used a double heating/quenching, triple temper heat treat. Cryo after each step. :o

08-29-2004, 11:10 AM
Hmmmm....well I can see where cryo after every quench might help but if he was cryo treating after tempering he probably didn't get much benefit from it. Tempering will cause retained austenite to transform to bainite or martensite, depending on the temperature. I can't see that there would be much retained austenite to work with the cryo after tempering. But then, this wouldn't be the first time that someone tried something that wasn't supposed to work and it did! :rolleyes:

08-29-2004, 12:26 PM

I tend to agree with you on this. My guess is that if he really was going through all these heat/cryo steps that that all he really achieved was perhaps a very slight gain from the double austenize/quench/cryo process and a big waste of time on the post temper cryo.

It is my understanding that a post cryo temper is needed to relieve the stresses from the deep freeze. If this is correct, then his post temper cryo would yield a weaker blade--not the tougher blade he was claiming. :confused:

08-29-2004, 12:40 PM
The post cryo-temper is to treat the new Martensite. If you do multiple temper/cryo steps, at some point it wouldn't matter any more.

09-11-2004, 05:50 AM
Cryo forms new martensite which is very brittle So you must ALWAYS temper after cryo......Many bladesmiths have the TRIPLE mantra !! Normalize ,harden ,temper ,cryo each THREE times whether it needs it or not !! :D

09-11-2004, 11:03 PM
Just a quick, semi-off subject question.

would triple quenching an air harden steel, such as D2 have any effect, good or bad on the finished product?

09-11-2004, 11:52 PM
I've done the triple quench stuff as part of an exhaustive, to me at least, experiment. I say get it right the first time, don't dance for the full moon, don't give a crap which direction the knife is directed toward the four points of the compass, and keep all your clothes on and save the chants for those who believe in that stuff. Except for that a good long prayer seems to help.

Forget the triples and learn to do it correctly in the first attempt (double temper though, double temper).

Okay, let's look at it this way: Paul Boss is king of the hill heat treater amongst the knife maker community. I do not say that sarcasticlly. I have as much and perhaps more respect for Mr. Boss as those among us that sing his praises the highest. Now let me ask you this: do you really think the best-of-the-best triple quenches yours blades? I will say this; if he thought it needed he would.

I'll tell you what; I'll be willing to eat crow if I can better myself by doing so. If a super triple quench advocate wants to we can do a test. A unbiased maker can forge, stock remove, or whatever two similar samples. One goes to me and the other to the triple quencher. We each then send the pieces to another unbiased tester. I don't care about the steel chemistry. It can be on the fringe of simple high carbon or S90V, BG-42, or any of those in between. I'm willing to hang my ass out in the breeze in front of all us. We must remember that this is about triple quenching. Normalizing and the number of normalize cycles has nothing to do with this. If I choose, depending, to normalize at all or more than once it has no bearing on the subject at hand. -- This is the sincerest form of disagreement I can come up with. Should I be proven to have errored I will humbly accept my new found knowledge and bow to triple quench.


09-12-2004, 02:04 AM
Normalizing and the number of normalize cycles has nothing to do with this. If I choose, depending, to normalize at all or more than once It has no bearing on the subject at hand.

Actually I think normalizing does have some bearing on it. If the test blades are in an annealed state, then the steel will require a longer soak at temp to get fully into solution then a normalized blade. Multiple quenches can compensate for not soaking long enough, so if your blades are annealed before hardening then multi-quenches might make a larger difference then if you normalize before hardening. Also since normalizing helps to make the grain smaller, then multiple normalized steel will get into solution quicker with the small grain then single normalized steel, unless your grain was already small to begin with.

09-12-2004, 02:10 AM
The subject matter is triple quenching. That is the matter in question. It is pretty simple and I did not dictate that I would normalize at all or at what stag it may be done if done by me. Let's not throw ringers. I have been specific and I will also be specific about what steps I actually took, should it come to pass. Its not a matter of who is correct but rather what is correct. Honestly, I don't understand your point as it pertains to this matter. Soak times are up to the quenchers. It makes little difference so long as one quench theory proves better than the other. Why, as one question, would my grain size before single soak disqualify my attempt?


09-12-2004, 03:55 AM
Okay AwP,

I went back and spent better time understanding your post. In the case you site I see what you say. But, in the light of what you say (and I am not arguing with your point here), how does it explain Ed's comment about certain quenches above three make it worse but then it becomes better again with an addition quench? I hope I remembered that statement right. My point in the original challenge above is this: I can do what I deem necessary in order to prepare the steel for single quench. Who ever else can do as they deem necessary before triple quench. Now, that does not mean triple quench and test and then quench some more until you get it right. Triple quench only. No more. No less. In my case it is the same. One quench only. Period.


09-12-2004, 11:57 AM
In my opinion, the only truley fair way to go about this, is for somone who is quite skilled at both single, and triple quenching, to do ALL of the heat treating. Have some one produce 2 blades as identical as possible, preferably stock removal. Then ship them to the heat treater.

The heat treater should do everything the exact same on both blades as far as prep. Same anneal, same normilizing. With the ONLY difference in the blades being that one is single quenched, and one is triple. In fact to elimnate the doubt about normalizing vs non normalizing, it might be better to have 4 blades. 2 non normalized, 1 of them single, one of them triple. The other 2 blades normalized, and 1 single and 1 triple.

That should eliminate enough of the variables to prove beyond a reasonable doubt which combination is the best. To better explain my statement I'll lable each blade 1,2,3,4 and describe there heat treat. This is assuming all blades arrive at the heat treater in the exact same condition grain structer wise, which is why I'd prefer to do stock removal if I were in charge of the testing. And assuming all recieve a good and identical annealing.

1. Non-normalized, single quenched.
2. Non-normalized, triple quenched.
3. Normalized, single quench.
4. Normalized, triple quench.

Out of those 4 blades the testers should be able to produce the most accurate and useful data on the subject. If anyone, or any group of you more experianced guys do this, you MUST post all the data.

Thats my opinion on the who testing procedure, should some one decide to give it a shot.

09-12-2004, 12:35 PM
And what criteria are we going to use to determine which is the better method? Hardness? Abrasion resistance? Toughness? Grain Size? Depth of hardening? Amount of retained austenite? I am a believer that each steel will respond to a "best practice" but different steels have different best practices. Personally, I don't triple quench, I do normalize once, sometimes twice. But I do not sell my blades for hundreds or thousands of dollars. If a person is honestly attempting to create the ABSOLUTE best edge possible, multiple treatments may have some benefits. However, I also think that one quench, and a double temper, on most steels will get you 90%+ of what multiple heat treatments will give you. If you want to add a cryo treatment, you can pick of a bit more.

I do not make my living bladesmithing. If I did, I would have starved by now. Trying to make a living bladesmithing in todays market requires you to differentiate your product from your competition. Some folks resort to mysticism, some to exaggerated science, some to the use of extremely expensive and exotic materials. I say "God Bless you one and all". You have chosen a tough way to make a buck so do what ever works for you! :D

09-12-2004, 11:49 PM
Quenchcrack, I believe that response deserves an "Amen!"

09-13-2004, 12:24 AM
I vote for Mixonknives idea. It takes me off the hook and besides, I'm biased. It also is better because it eliminates any doubts about one heat treaters abilities apposed to the other. So if one fellow does both and he is not biased to begin with it would be much better. For the proceeding post, grain can be visually analysed by those that are accustomed to doing that. Edge retension can be tested. Retained austenite would require an outside source to determine.


09-13-2004, 01:01 AM
Sorry it took me a while to get back to this...

how does it explain Ed's comment about certain quenches above three make it worse but then it becomes better again with an addition quench?

I have no clue, that one is way beyond me. :confused:

My main point though was just that normalizing can make a big difference besides the single/triple quench, so it should be done the same on both test blades.

I vote for Mixonknives idea.

Me too, I think he gets what I was saying and explained it alot better, rather then just using examples without explaining my actual point like I did.

I nominate 52100 as the test steel. It's the one that the most people claim beneficial effects of triple quenching with, so if triple quenching has anything to it then it'll be the most likely steel to show up noticeably under testing.

09-13-2004, 08:42 AM
You should add 5160 to the test as it is smiler to 52100 and also mentioned as benefiting from a triple quench. Gib

09-30-2004, 10:02 AM
I have already done almost this exact experiment. I posted the results over on the "other" forums in a thread started by "pig". I cannot get the search function to work so I cannot post a link.

Roger, maybe you or someone else could search it out and post a link here.

I am in the process of moving so I don't have my notebook handy but the test I did used L6 steel from a saw blade and went something like this;

I cut two blades from the same piece of steel using a plasma touch. I then cleaned them up and profiled each to the same style of blade. I did one normalizing and two spherodizing heats on both pieces. The blades were then ground to about eighty percent and hardened. For heat I used my Paragon heat treating oven. The austenitizing temp was taken from the crucible manual along with soak times. Both blades were put into and removed as close to the same time as was possible for me. Both blades were removed and quenched at as close to the same time as was possible for me. they were let cool to the temp of the oil (165 degrees) and then removed and put in cool water and then to the freezer to cool to -20 nd then into Ln2 until they stopped bubbling then into sifted wood ashes to warm garadualy. At this point I took one blade and tempered it and the other I repeated the hardening process twice more. Each blade was tempered three times for two hours at 350 degrees. Bothe blades were finished to an a30 finish and sharpened on a Norton fine india stone. testing cosisted of an edge flex to test for proper hardness, cutting hemp rope to test for edge retention and a 90 degree flex to test for toughness.

In the edge flex (brass rod test) the single quench blade made several flexes but chipped while the triple quenched blade was flexed more times with no chipping.

The single quench blade was tested for edge retention five times and averaged. the triple quenched blade made as many cuts without sharpening as the single quenched blade did over five sharpenings.

I the nintey degree flex the single quenched blade broke at about eighty degrees while the triple quenched blade bent to nintey and returned to straight on its' own.

I left out the specifics of the tests because I could not remember the exact numbers. The blade shapes and geometry were those of a boning knife as would be used in a butcher shop. I did my best to keep all things equal in the treatment of the knives during the making and testing. I did not keep track as to which knife would be triple quenched until I started the second quench on the blade that I picked. And last but not least I, to the best of my ability tried to keep any kind of bias out of my tests. some of the steps preformed were done simply because that is the way that I do things. (ie oil temps and double spherodizing heats.

One more thing. I am not critisizing anyone or telling anybody how they should be making there knives. if a person is happy with a single quench and temper good. If they want to do a double quench that is ok with me. I belive that I get the best performance out of the triple quench and triple temper. I relate the above test only for the pupose of telling about what I have done in my shop.

Bill Burke