Triple this, Triple that... what gives?

69_knives

69_knives

Joined
Nov 20, 2008
Messages
1,855
Reading a lot about people doing each heat treat step three times... Normalizing, Hardening??? and tempering.

Is all this really necessary?

I come across this

which graphs some of the differences in properties of steel with different carbon contents.

I'm still going with the 'knowledge' that my leaf springs are 5160 and on each of those charts, there is only a small difference in properties for a steel with 0.6% carbon between the annealed and normalized state when compared to a steel with 1% - 1.2% carbon.

That data doesn't state that multiple normalizing was done so in the case of 5160, I'm going to have to surmise that once is enough.

Now, what of the triple heat/quench and triple temper. Is there particular data that can back up the benefits that somebody can point me to?
 
I am not a metallurgist, but, I believe that I have learned a bit in the past few years and this is what I believe are the reasons. Someone who understands the process better please correct me if I am wrong. The triple normalize using a descending "hot" temperature will help insure that your steel is back to a small grain size from any overheating during the forging process. I personally never triple quench. I think that sometimes it is done to accomplish about the same thing as multiple normalize cycles. Triple temper might be a bit much for 5160, but not one with a lot of carbon like D2. But, a double temper is good. Here is why. When you harden your steel a bit of the steel will be retained austenite after the quench. When you temper the first time much of this will convert to martensite. But, it will be untempered martensite, so you need the 2nd cycle to temper this additional martensite. A steel like D2 with 1.5% carbon will end up with loads of austenite. You can convert much of it with a cryro treatment or you will need the third temper as each of the first 2 will convert some of it to austenite. You may never get it all converted but, at some point it is diminishing returns.
 
Well I just want to set the record straight on this issue, If you heat treat and temper the blade right the first-time, there is no need to repeat the process the second or third time. And that's the bottom-line. :thumbup:


Terry
 
Last edited:
Oh boy!, I kind of told myself I would just leave this topic alone in the future, but I like the way Mr. 69 asked for input and he deserves some reasonable explanations along with what he may hear elsewhere. So if there are any zealots waiting to pounce, hit me with your best shot but I doubt I will learn to keep my mouth shut any better after the assault;).

To some extent I agree with the replies so far. Much of what is accomplished with triple quenching can be done in the normalizing stages. Folks working with steels beyond their tools ability to properly treat may get a boost from doing some things more than once, if they are careful. Much of the perceived benefits of an exaggerated nature from multiple treatments are simply that- exaggeration or more often misinterpretation of cause and effect. A 20% increase in any property over the controlled baseline is interesting and perhaps worth pursuing, 100%, 200% or more, stretches credibility and is most likely error or slick working of the numbers for marketing.

Now with that being said, a closer look at each is warranted.

Triple normalizing- this can have an effect on the steel in a couple of different ways the two most prominent would be grain size and carbide condition. If you have heavy banding or other segregation in a high carbon steel, heating it above Accm (the upper critical temp for a steel, which is much hotter than hardening temperatures) will dissolve the current carbides and put everything into solution. Quickly cooling it in air from this condition will keep it all evenly distributed and fine, but it will not help with grain size, in fact it will increase the grain size a bit, if this treatment is then followed by a couple of cooler heats the grains can then be recrystalized into a finer condition. This is the big selling point of the triple quenching crowd, but normalizing will do it with much less stress on the steel and without buggering up your finished blade in the final sequence. Also bear in mind that there is a reasonable size for grains in most steel applications and taking it farther just to boast about the finest grain around is merely overkill and can even have some negative side effects. So refine the grain to a good working size and move on. The real challenge in proper normalizing is in the carbides, grain refinement is rather elementary since it happens every time you heat and cool the steel at a reasonable temperature.

Triple quenching- more of a marketing buzzword than anything else, in my humble opinion. It is basically doing much of what should have been done with good normalizing, but can push grain size down much faster due to increased stress on, and strain within, the steel but all of us know what this also leads to when it gets out of hand. Many of the wild claims attached to it that are not error in testing can easily be attributed to the elimination of other problems that were not taken care of in earlier heating operations. Guys who are working with carbon rich steels that require accurate and extended soaks may find that repeating the austenitizing and quenching procedure will allow them to put much more carbon into play without a proper soak than if they only did it once. It is in this part that the sayings about “doing it right the first time” apply the most.

Triple tempering- this one has some merit as well, but in a very technical sense if you could it all right the first time the other two wouldn’t count here either, unfortunately even with the best of heating equipment we cannot accomplish everything that would be best for the outcome in one heat. Almost any steel above .85 carbon or having added alloying will tend to retain some austenite, many of the ones we bladesmiths work with have it in such small percentages that it is inconsequential. When we temper a freshly hardened piece of steel we allow the carbon to move in very small and controlled ways, when the leftover austenite loses carbon in tempering it becomes less stable and can kick over on cooling. Now you will have a tempered blade with tiny bits of fresh untempered stuff that just kicked over, so the next tempering will cure this. There will also be a switching from one atomic stacking to a bit more stable one, and time at temperature seems to be a finer tuning than just higher temperatures so your results can be better by spending more time at it instead of just going for hot, hot, hot.

For what it is worth I keep a log in my shop of every single thing that ever goes under my Rockwell pentrator, in hopes that one day that mass of data will be handy. The one constant observation that I have made is that multiple temperings seem to have a very nice “evening” effect on the readings. As quenched I may get a range from 65HRC to 63HRC on a piece, but after the second tempering the range will narrow to one point of deviation, and on the third the entire piece will give me almost the same reading (less than .5 HRC).

I triple normalize (or however many time sit takes). I triple temper because it is nice to walk in the exact HRC you want while gradually bumping up the temperature, and I cannot ignore the equalizing effects. I do not triple quench, because I feel I accomplish what I need in the normalizing stages and then keep it there with the specific anneals I use. I do not have a problem with people triple quenching if they like in order to get down what didn’t happen in the normalizing, but do I have to admit to having a problem with people making some of the rather shameless claims about it if they do. The operations I have seen a benefit in will give a moderate increase in some properties that I like, nothing supernatural or even all that noticeable to the average objective user, but it gives me that little extra satisfaction:)
 
Last edited:
I triple normalize and temper because I get the best performance from 1084 using my current set up in tools. When I get different tools/steel I will work on a new process to optimize cutting at that time.
 
Tempering - simply put the simple steels [low carbon and alloy] require 1 temper ,the more complex steels 2 tempers and the most complex steels require 3 tempers.
In all the HT proceedures you should never do things without knowing exactly why you are doing it .What are you starting out with, what are you trying to accomplish, how should you proceed !
When you've done it right you can sit back and relax with a glass of Triple Sec !!
 
Here is what I have experienced on topic:

On an another thread I consulted to recover my annealing error for my D2 blades. I triple normalized as Kevin and Mete suggested to refine the mess I had in steel. Now 2 of 3 blades are sharpened and I tested them, file test, brass test and hard using etc. They out perform all my previous D2 blades. The hardening temperature and tempering method, geometry is similar if not the same but the triple normalized ones are exceptionally better than just hardened stock removal blades. I suspect the D2 comes not fully annealed or overheated from factory... Even for stock removal and high alloys some normalizing cycles may guarantee to have the best out of that steel...
 
Can I add a question here that is probably simple but my finals ridden brain can't think well right now- Why is is double/triple temper for an hour rather than a 2 hour soak? All I can think is decarburation but at the lower tempering temperatures I would guess it would not be much of a concern.
 
Well, "Mete's" 1,2,3 temper for different complexities of steels is for 2 hours each. It's a time thing (at a temperature) in that it takes a while for the carbon to move and 2 hours is how long it takes.

Mike
 
How about a little Rumor Control Dialogue as my teacher in high school used to say? I have heard that you can actually overnormalize simple steels with no carbide formers like 1084 to the point where the grain size is so small that they are more difficult to harden. True or false?
 
Reducing grain size changes hardenability so after three normalizings you're losing ground !
During tempering the temperatures are too low to produce decarburization or scale.Multiple tempering is there for a reason. First temper reduces retained austenite .On cooling however this means freshly produced martensite is UNtempered and must be tempered.
Tempering is a time and temperature controlled process .Two hour tempers are a good time.
 
The metallurgical facts on this thread are all twisted and confused! LOL :D

Let me try to straighten it out…
Simply put,... you only need to “accomplish” each step once. Quenching is to harden the steel. Normalizing is to stress relieve the steel, (resulting in a soft homogenous fine grained pearlitic structure). Tempering is to stress relieve the steel and lower the hardness resulting in a tougher blade.

If quenching or normalizing cycles are used strictly to refine the grain size then they are used as two different forms of “thermal packing". If a single tempering cycle results in some un-tempered martensite then the steel isn’t completely tempered yet.

So,… regardless of how many cycles are used, you only need to accomplish each step once.
 
Last edited:
Tai Goo said:
The metallurgical facts on this thread are all twisted and confused! LOL :D


So,… regardless of how many cycles are used, you only need to accomplish each step once.
Click to expand...


Now that you put it that way, it makes a lot of sense.
 
With the "normalizing" and "thermal grain refinement" cycle or cycles, it really just depends what state the steel is in, prior to, or going into the quench. In other words if it already has a refined grain and is stress relieved, or prepped for hardening whatever the case may be,... no more cycles are necessary. On the other hand, if it has an abnormally large grain and is all stressed out, it may take several cycles... Like Kevin said,… how ever many times it takes.

But... don't “abnormalize” it. :)

I always think it's a good idea to get each step accomplished in as few heats as necessary to get it right.
 
Last edited:
CUTS LIKE A KRIS said:
Can I add a question here that is probably simple but my finals ridden brain can't think well right now- Why is is double/triple temper for an hour rather than a 2 hour soak? All I can think is decarburation but at the lower tempering temperatures I would guess it would not be much of a concern.
Click to expand...

Also remember that in an oven it takes a good percentage of an hour just to bring the blade up to temperature and then only what’s left is used to actually temper, thus the 2 hour thing is a good idea. I do 1 hour or less in my first temperings because I am working with salts. I just saw an episode of “Good Eats” with Alton Brown that demonstrated this better than anything I have ever seen. He had a half dozen little ducky ice sculptures and tries to thaw them in several different mediums. One that was in the hot oven was not much more melted than one put in the refrigerator, while the one that was on cool water for the same amount of time was liquefied. This is the power of a liquid to conduct heat over that of air… insulator/conductor. Thus I must be careful not to pass along my tempering schedule times to folks with ovens and kilns or they will accomplish very little.

I have read, and experience has confirmed, that longer times at lower temperatures will give a better toughness at higher hardness, then shorter times at higher temperatures.
 
In the long run,... I think it takes some common sense and good judgment.

Speaking generally, it really all depends on what you want and how you want to do it under any given set of circumstances or parameters... It depends what you are making. For example with flint strikers and striking surfaces on striker knives, I actually grow the gain and leave it dead hard so it chips out easier and throws off a larger spark.

Another thing is that heat treating is an integral part of hot forging. In other words the steel is “heated”, primarily to soften it and make it more malleable under the hammer. However, the heating and cooling used during the overall forging cycles may be simultaneously used for other thermal advantages…
 
jdm61 said:
How about a little Rumor Control Dialogue as my teacher in high school used to say? I have heard that you can actually overnormalize simple steels with no carbide formers like 1084 to the point where the grain size is so small that they are more difficult to harden. True or false?
Click to expand...


This post will simply support every thing that mete said, but I thought I would give some reasons for the phenomena. It is all about how diffusion gets started. Soaking to make austenite is a process of diffusion, nucleation, formation and growth of new austenite grains is a process of diffusion, and definitely the formation of pearlite is a process of diffusion. Still one of the best analogies I have for it is the surface of a Michigan pond this time of year, the ice forms first at the edges and around sticks or rocks disturbing the surface, this is were the ice crystals first nucleate and then work out from there. The more sticks or irregular edges you have the more directions the ice forms from, so it could be seen as many different ice sheets patches growing to cover the pond.

If you want to increase hardenability you eliminate as many points of nucleation for the pearlite as you can, i.e. larger ponds with no debris and very smooth sides take the longest to freeze. So when your steel doesn't want to get as hard as you like, you can bump up the soak temperature and dissolve few more carbides (sticks and rocks), or perhaps even increase grain size (less pond edges) and this will retard the formation of pearlite on cooling because of the fewer places for it to begin.

Now the opposite. If you cycle you steel to get finer and finer grain, you will also multiply the amount of grain boundaries with every cycle thus making it easier and easier for pearlite to form and lowering hardenability. Low temperature cycling can also bring carbon out of solution and make more sticks and rocks in the pond. A simple test anybody can perform to confirm I am not full of $&!* on this is to quench a bare blade of 10XX steel in a medium to slow speed oil and then etch it to observe the natural hamon. Then if you refine the grain and do another heat and quench the hamon will move closer to the edge with every cycle due to the diminishing ability for the steel to harden to a give thickness. But if you heat it once to a higher temperature or use a longer soak the hamon will jump back near the spine. This is just one reason that we need not be too much in awe over anybodies ability to play with grain size since it is quite easy to do once you understand it.
 
You must log in or register to reply here.
Back
Top