Triple quench secrets to be revealed?(or not)

Status
Not open for further replies.

pig

Joined
Mar 12, 2003
Messages
100
Can triple quench secrets to be revealed? Yes, for sure sometime.
Am I revealing the secrets now? I do not know, just giving a nice try.

To be understood:
(1) I am not a native speaker of English, sorry about language mistakes.
(2) I am not a metallurgist, just have read some books for fun.
(3) I am only a hobby smith, making a "brain gym".

The triple quench mystery is especially interesting (for brain gym), because it is not only one unsolved problem ??? (I mean, reasons why it works) but also seems to have a logical contradiction.
On the material world there are not, however, such things as logical contradictions (I am not talking about the world of ghosts or spirits)!


(1) Ed Fowler's tests show triple quench works, and is reliable.
(2) Metallurgists say it can't work, steel has no memory and if we quench many times it can't be better. And they have their own proofs.

Here we have seemingly a logical contradiction. So, we just have to find out why really the knife is better when quenched 3 times even the hardening (changing austenite to martensite) can't to be explained to do anyting like that.

"Seemingly" is the key word.

I have worked one lifetime with computers and did learn that always when I am saying "this is impossible, can't work like this", there is a perfectly clear explanation just to be found.

So, there is some odd factor x, which happens during the triple quench process (not to change austenite to martensite). Which is it?

Now, to the factor x:

Let start with fashionable "super steels". (I am a hobby smith and do the work for the old tradition. I am not very interested on fancy alloys, which demand some fancy ovens and so on.) However, we can get the matter in the question more easy to understand with an example.

Mill specs about fancy steel x tell that you must keep the temperature y (high) for z minits (very many) before quenching. Nobody says: quench immediately when temperture is y.

Now, we can explain why we must to keep long time the heat. Many fancy carbides as Vanadium carbide are very reluctant to let out carbon as needed. They need lot of time and heat.

Unfortunately also ferrite carbides (carbon steel and low allow steels) need time (few minits) to let out some carbon as needed. (A low carbon steel is the most critical) So, basically the carbon steels are similar to fancy new steels, the keeping time exists, but it is much shorter.
A short (for carbon steel) proper keeping time do not enlarge the crystalline structure!

More error prone thing with carbon steel is to overheat it. Smiths are for this reason advised to quench "immediately". In practice a beginner smith always needs some extra minits to get long blade evenly heated and is satisfied the color all the way, and gets enough carbon out by accident.

Now, if we are skillful enough to heat optimally and really quench immediately we need 3 times to quench (and to heat for carbides) to provide the needed "carbon let out time" for optimal result (4 or 5 times even better ?). (Between the quenches the carbon do not re- enter into carbides.)

One more fact on favor my theory: Ed Fowler says that the steel seems more soft to sharpen after 3 quenches.

The hard ferrite carbides disappearing! Wear resistant carbides are also making knife more difficult to shapen (Cr-carbide HV 1600, V-carbide HV 2800 (HRC 85!))

If somebody want to test this (A joke, nobody will!), the difficulty is not to overheat because overheating is even more critical thing.

So, triple quenching protects from overheating and works, on the other hand it is cubersome.


pig from Finland
 
Interesting.

I was talking with a Master Smith recently. He uses an oven on 52100 (Fowler-steel). Says that unless he gives it at least a 5 minute soak, he doesn't get the results he wants. Kinda jives with your theory.

I've just finished a 52100 knife with a 5 minute soak and was going to start destruction testing on it tonight.

And considering how frustrating work is right now, it should go pretty fast.

Steve
 
A metallurgists response. I have never understood the triple quench and in my industrial experience it is never done. Here is the heat treating process - typical knife steels are sold in the spherodized annealed condition, otherwise they would be very difficult to machine. When it is heated to the austenitizing temperature the spheres of carbide start to dissolve. This process takes time ( called soak ) and continues until the austenite is saturated with carbon ( about 85% C), any remaining carbides will stay that way and give the blade wear resistance. Insufficient soak time will not produce optimum strength martensite and excessive soak will cause grain growth which will weaken the steel.Insufficient soak time might take three tries to dissolve the carbides but why not do it right the first time. Are some knife makers paranoid about grain growth ? Grain growth takes time too, some alloys more than others ,but a proper soak is finished before there is any grain growth problem. So normalize once (proper above critical temp not subcritical) ,harden once, temper well maybe more than once .
 
and a proper ramp speed and soak period allows the entire steel cross section to become consistant in temperature. RL
 
pig,

You'll find quite often that knifemaking has many means to the same or similar end. Though not all are balanced in their efficiencies, the end results can often make a blade that is good regardless. Occasionally makers will cite the, "if it ain't broke don't fix it" adage for their craft as a rule of thumb and engage in research no further. As long as they have what they want. While others, through experimentation and endless study will eliminate wasted or negligible steps in an effort to arrive at "optimum" and "efficient" bladesmithing progress and skill.
 
I have used 52100 for several years now, and I agree that soak time is critical in getting full hardness in the quench. Per Paul Bos, a 10XX steel needs only reach critical temperature for a moment and when properly quenched it will reach max hardness. When you add any alloy (like the chrome in 52100) you create a situation where the soak becomes very important.

I, too believe that a 5 minute soak is the ticket in an oven for 52100. A few months ago, before visiting a friends house, I made 4 test knives of 52100. 2 were triple quenched, and the other 2 were single quenched. All 4 were heated in an over with a 5 minute soak prior to quenching. When I say soak I mean that the blade reaches oven temperature and then the 5 min clock starts. The results showed that all the blades performed identically in cutting rope. We really expected to see a difference and when our fingers were so sore from cutting, we hit up several of his employees to see if they could tell a difference and none could.

This leads me to believe exactly what "Pig" theorized. If you heat evenly and soak through for the appropriate length of time, you will get max performance no matter how many times you do it. When Ed Fowler does it with a torch, he is not able to fully soak the edge and the triple quench will improve over a single quench. Ed has his own reasons for doing it with a torch, and those are not in question here.

What this all goes to show is that you cannot try to apply some of a maker's techniques to your own and get the same results as him. Every individual factor in the creation of the knife plays a role in the outcome. It is therefore inappropriate to make blanket statements like "triple quenching is better" etc., without considering all the unique factors that affect the outcome.
 
Burt,

What this all goes to show is that you cannot try to apply some of a maker's techniques to your own and get the same results as him.

Well said. Don't know how many times it's happened to me. Probably why it takes so long to develop something for yourself. Try as you might to duplicate someone else's work, there's always some little variable you didn't know or think important.

Steve

PS just got done testing my first successful 52100 blade. Used a 5 min soak @ 1550. Held up thru 33' of cardboard and only took one pass on the buffer to get the edge back.
 
So there you are. Temperatures, soaking times, and rapid quenching. Per different steels and thicknesses the first two can vary and a few multiple quenchings may make up for misjudgements or inconsistancies in previous HT sessions. In no part of any HT tech sheet have I yet seen a recommendation for multiple quenching. It does make sense though that under open atmosphere conditions un-even heat transfer could necessitate a justification (or need) to double or triple quench. And, under that type situation I have noticed myself a marked improvement by double quenching.

RL
 
Burt, It's nice to see you on here sharing your knowledge--thanks.

A couple of questions: what sort of oven are you using? I ask because some have had problems with their ovens overshooting the set point when ramping up. Any experience with this?

Also, what is your quenching medium and procedure (tempering, too)? This must also have a great effect on the end product.

Thanks again,

John
 
John, I found by pure luck that my Even Heat over shoots when set to ramp quickly. It makes sense, I suppose, that it would in that situation. The over shoot appears at the lower temperatures, tempering ranges. After that the temperature begins to raise at a slower and slower rate. For equalizing and austenitizing it is not a worry with my oven. For tempering I have to be careful to let her level off before putting the steel in.

RL
 
John,

I am using a BGF7300 model 2. Actually, it's a homemade oven. It's propane with 2 venturi style burners in a 10" dia. tube. I have an analog pyrometer that's very accurate as far as I can tell. I actually need to build another oven with some improvements, but this one works fine. Becuase it's gas and the heat is "instant" as opposed to electric, if it's moving up too fast you can shut down the heat and let it sit, and then fire it up again. I have an inspection hole in the front door that I look through after the oven is at temp (1475 deg), and wait to see the whole blade go critical. I then start a 5 minute clock.

I quench in 120-130 deg. oil that's a 50/50 mix of motor oil and automatic transmission fluid. One nice thing about that type of oil is that trans. fluid is a fast oil and motor oil is slow. By mixing the two, you can get and oil that works just at the right speed for what you're doing. Yeah it flames up and stinks, but that's part of my look.
 
I am glad to get positive comments!

Indeed, I did take the question not mainly of the point of view of the practical smith work but as a nice challenge to solve something which seems to be a logical contradiction (read impossible).

People are arguing who is right and who is wrong with this contradiction.

In fact nobody is wrong!

If we do quenching as Ed Fowler (immediately) and get 100% guarantee against grain growth we must quench many times, and also get spine not heating.
Also we get this soak time (soak time is new word for me).

On the other hand metallurgist are right knowing that if we change austenite to martensite (quench) and do it again the steel do not remember that it ever earlier was martensite before. This way metallurgist's doubt is more than understable.


Good quenches triple or single: pig
 
This is a very interesting thread. I can't help but wonder if there
is any difference in the amount of decarb between single quench with
a 5 min soak and multiple quench. Many years ago I worked in a lab and
we used argon gas for an inert atmosphere. A ceramic tube was placed in the furnace and the argon gas flowed through the tube. No O2, no oxidation! We also used a lot of Hydrogen, but that's another story.
 
Well, I am talking too much!

I want to thank specially:

-Burt Foster
-itrade
-Don Powel

making me feel useful.

What I tried was to pass something I feel imoportant, I have learnt hard way (not only connected to triple quenching).

(1) It did take me tens of years to really give up "mystery thinking" with computers. I think it is a human basic behavior to accept mysteries. It takes time to learn such a simple thing that in techical or material world there are no such things. (Totaaly different to wordl of ghosts). You have not just done your homework, taken a more bread view.

(2) People have wrong idea about human commnication. If you say something you do not understand that other people have their own different filter to interprete it (As triple quenching and metallurgists. If you name triple quenching as triple heating the end result is different.)


pig
 
Thanks Pig, for a very interesting subject. I have wondered about this myself.
All the responses so far, seem to agree that 52100 needs to be held at critical temperature for 5 minutes before its quenched. Whether this is five minutes at critical temperature and then quench, or three quenches that add up to five minutes at critical temperature doesn't seem to matter.
My question is, would reaching critical temperature and quenching three times add up to a total of five minutes at critical temperature?
When I made my ABS Journeyman test knife I heated 5168 to critical temperature and quenched once. The knife easily passed the test. If I had heated to critical temperature and quenched two more times, it would not have been close to a total of five minutes at critical temperature.
I don't know about others, but when the edge of the blade becomes non-magnetic, I quench.
To reach a total of five minutes critical temperature quenching three times, it seems that the blade would need to be held at critical temperature for almost two minutes prior to each quenching. Does anyone do this?
 
It was an interesting thread Pig. There are many things in knife making that I don't understand why they work. The triple quench is one of them and in those cases I bow to the experts like Mr. Fowler. The simple fact is that we have so many experts here, it's difficult to keep up with the information flow.

Do you have a web site? I have a special interest in the knives of your country!
 
I have wondered about this to, I have read where 100+ years ago some Smiths Knives were a lot better, way above average in edge holding. What thy did seems to be a mystery and I wonder if there process died with them and was not passed on to younger generations. The question is this one of them? And was rediscovered by accident.
Gib
 
peter nap:

Do you have a web site? I have a special interest in the knives of your country!

I did make a search with words (puukko=knife and seppa= black smith). I got 400 hits. (In fact the word is not seppa but dots on top of a, a letter you can't find in US keyboard.)

Because the text is often in Finnish, pictures can be useful. So, you can start with (Galleries):

http://www.puukkosivut.net/4images/

Some links to makers:

http://www.puukkosivut.net/links/LinksCaffe/links.php?cat=10

At least Roselli has some English text:

www.roselli.fi

He has flags for languages. No US flag!
Perhaps UK is ok?
(Excuse me supposing you can't read easy Finnish language! Even small children here can read it.)

I have website of my own: http://www.evitech.fi/~tk/

This site is tells more about other things than knifes. (For example small operating system kernel for computers written by me. The knife part is not up to date.)


Have a nice trip to Finland: pig
 
All the latest writers do not seem to accept my arguments. It is understable. Finally the case is that we would need a similar kind of conrolled test as Ed did for soak time to be sure!
At least "itrade" and Burt Foster referred to pracical results and experiments earlier in this tread.

There are at least 3 things making this difficult:

(1) Mill specs do not in general tell the proper soak time or lower and upper bound of it. Is the lower bound 2,3,4 or 5 minits for 52100?

(2) The soak time is very critical only for high alloy steels.

(3) Soak time is again less important to steels with high carbon content. The steel do not need the carbon dissolved example from Cr carbides because it has enough carbon anyway. (Read the answer of the metellurgist "mete" earlier in this thread.)

It is important to understand that steel can't use to create martensite any carbon more than there is in 0.85% carbon steel. Rest are staying to make wear resistant carbides.

Ed Fowler's initial controlled test was done with 5160, which is wery low carbon steel, about 0.55% C. For this reason soak time is more more imoprtant than in most cases!

If we redo test with different steel it is a different matter.


pig
 
Peter, I bet if you asked pig real nice he'd send you some pictures of his incredible knives. He is very modest, and one fantastic knife maker.

Dave
 
Status
Not open for further replies.
Back
Top