Fine Tuning a Heat Treating Protocol

A

anonymous76

Joined
Feb 10, 2019
Messages
19
Well here goes... I apologize if a thread about this has been started elsewhere (I looked). I realize this may be opening a big can of worms, but I am curious as to how more experienced bladesmiths went about optimizing their heat treating protocol for a given steel. For sake of argument I am thinking about 52100 since several protocols have been proposed in this forum. The stickies provided in this forum by Stacy and Larrin's recent article were very informative. These links seem to be a good place to start, but there was little consensus between different recipes. With so many variables to test and fine tune this seems like a daunting task. Where does one begin? What is most important? How do you test your blades, assuming you do not have a laboratory at your disposal? First off, I would not attempt a low tech heat treat on such an alloy steel, so sufficed to say I have a Paragon HT kiln. The first, obvious step before applying any of the above mentioned protocols would be to calibrate the thermocouple. Before I get ahead of myself, perhaps it would be best to define the intended use of this steel. I do not believe there is one perfect steel for all applications. For a relatively rich forging steel such as this I personally would not make blades over 5 inches. For larger blades that take a beating I would probably use 5160 or L-6. Below is a short list (not comprehensive) of the variables that may or may not be critical:

Steel Supplier
Initial State of Steel
Forging Temperature
Forging Time
Normalization Temperature
Normalization Soak Time
Thermal Cycling (grain refining) Temperature(s)
Thermal Cycling Soak Time(s)
Number of Thermal Cycles
Annealing Steps
Annealing Temperature(s)
Annealing Soak Time(s)
Annealing Cooling Rate(s)
Degree of Finish Prior to HT
Stress Relief Temperature
Stress Relief Soak
Austenizing Temperature
Austenizing Soak
Oil Type
Oil Temperature
Time in Oil
Number of Hardening Cycles
Cryogenic Soak Time
Tempering Temperature
Tempering Time
Number of Tempering Cycles
Time between HT Cycles

Some of these, such as steel supplier, may be treated as constants (at least initially) but which? Finally, how did you determine when the end product was good enough to pass on to customers? I realize this is a lot.
 
This should be good. I’d like to get, or put together, the same info for 1095 steel, and then 1095/15N20 laminate steel.
 
Wow this could be involved. For myself with aebl I made coupons and treated them at varying temperatures. I used to my ames hardness tester to get the highest hardness possible. I also broke each coupon to check the grain. All plate quenched. Then by messing with temper temperatures I was finally able to hit 64 to 65 rc from which I could temper back to what I wanted. I was initially tempering too high.

I guess with oil hardening steels you have to think about the oil temp. I the the rest may be over thinking it. There's a lot of info out there already to use as a starting point.

I do think you need a way to do hardness testing to compare results with some sort of measurable data.
 
tkroenlein, I think I know what you are talking about. Some of the HT protocols given in this forum mention a stress relief after grinding which may sound like annealing. If i repeated anything else, my bad.

Larrin, I expected you to respond to this. Thanks. I fully intended to use your protocol to the letter, initially. Basically making that as the blade to beat for further testing. If all protocols on this forum were the same, I would be less likely to wonder about the numerous variables that affect the steel. I don't expect that I will diverge far from what you set down in your article. Several key variables are of interest to me. For instance there is the quality of the material and its initial state. Both of these would vary according to source. To start off I was going to go to NJ Steel Baron. But how would 52100 from this supplier compare to that of others? And then there is the question of the oil used. One thing I plan to look at is hardening response of 52100 with varying oil viscosities (all other things being constant). I would start by comparing blades quenched in Parks 50 to that of very low viscosity mineral oil (i assume fast), and then gradually work my way up in viscosity from there. I could go on and on with all the thoughts going around in my head about this. I guess this comes with being a former geoscientist. Having constant curiosity and always questioning.
 
anonymous76 said:
tkroenlein, I think I know what you are talking about. Some of the HT protocols given in this forum mention a stress relief after grinding which may sound like annealing. If i repeated anything else, my bad.

Larrin, I expected you to respond to this. Thanks. I fully intended to use your protocol to the letter, initially. Basically making that as the blade to beat for further testing. If all protocols on this forum were the same, I would be less likely to wonder about the numerous variables that affect the steel. I don't expect that I will diverge far from what you set down in your article. Several key variables are of interest to me. For instance there is the quality of the material and its initial state. Both of these would vary according to source. To start off I was going to go to NJ Steel Baron. But how would 52100 from this supplier compare to that of others? And then there is the question of the oil used. One thing I plan to look at is hardening response of 52100 with varying oil viscosities (all other things being constant). I would start by comparing blades quenched in Parks 50 to that of very low viscosity mineral oil (i assume fast), and then gradually work my way up in viscosity from there. I could go on and on with all the thoughts going around in my head about this. I guess this comes with being a former geoscientist. Having constant curiosity and always questioning.
Click to expand...
Just start with that DET protocol and Parks 50. The NJBS did fine.
Grab a Dewar.

Grind the blades and test. See when the edges roll or chip at different Geometries and go from there.
 
Larrin said:
NJSB 52100 is referenced in annealing Part 2 when comparing as-received to DET hardening response: https://knifesteelnerds.com/2019/06/17/annealing-part-2/

Parks 50 is fine.
Click to expand...

I seem to remember that NJSB 52100 was never tested without normalizing first. I tried to increase soak time at 1475f to 30+ minutes about 5 years ago, and the steel wouldn’t harden. It needed the higher normalizing temperature to break up the course carbides.
 
Willie71 said:
I seem to remember that NJSB 52100 was never tested without normalizing first. I tried to increase soak time at 1475f to 30+ minutes about 5 years ago, and the steel wouldn’t harden. It needed the higher normalizing temperature to break up the course carbides.
Click to expand...
We got it hard, but not to the level of the re-annealed steel.
 
Willie71 said:
I seem to remember that NJSB 52100 was never tested without normalizing first. I tried to increase soak time at 1475f to 30+ minutes about 5 years ago, and the steel wouldn’t harden. It needed the higher normalizing temperature to break up the course carbides.
Click to expand...

The AKS without normalizing, cycles and annealing was still softer than the NJBS with.

So I feel best practice will still be to use that DET protocol regardless.
 
Larrin said:
We got it hard, but not to the level of the re-annealed steel.
Click to expand...

I remember I had good hardness after heat treat, but after two tempers, it was below Rc50, iirc. I’d have to find the old thread.
 
scott kozub said:
Wow this could be involved. For myself with aebl I made coupons and treated them at varying temperatures. I used to my ames hardness tester to get the highest hardness possible. I also broke each coupon to check the grain. All plate quenched. Then by messing with temper temperatures I was finally able to hit 64 to 65 rc from which I could temper back to what I wanted. I was initially tempering too high.

I guess with oil hardening steels you have to think about the oil temp. I the the rest may be over thinking it. There's a lot of info out there already to use as a starting point.

I do think you need a way to do hardness testing to compare results with some sort of measurable data.
Click to expand...

Was the 64-65 befor or after the first temper?
 
You need to add shape to your list of variables. Cooling is driven by surface area, but volume determines the thermal mass.
You also need to define what you want from the blade. Does it need to be 63rc? How tough? How are you measuring this, with what margins of error?
A lot of those variables you can just set at a known decent base value, and then ignore them until you need to find the last marginal gain.

Personally i have more interesting things to do than chase 0.5rc or a ft/lb of toughness. If i can grind it to around 0.1mm with out chipping and it holds a nice edge i call it good, but i want a hands on process i do by feel and i'm willing to sacrifice precision to get that.
 
Just a comment:
I feel you are overthinking this for knifemaking. Most of those variables don't vary much at all with a standard HT protocol for a knife blade. It would be better time spent learning to get consistent results with whatever equipment you have than chasing unicorns looking for the "perfect" HT regimen.
 
What I did with 52100 from NJSB was take 6 coupons, start all of them with the DET protocol and then divided them into 2 groups=> Parks 50 and McMaster Carr med speed oil quench. In each group I did aus temps of 1500, 1525 and 1550. In my kiln I found that 1525 into McMaster Carr med oil gave me the best results (RC66-67) as quenched. If you're hardness testing make sure you get all the decarb off as it will greatly affect measured hardness.
 
Here's an excerpt from Larrin Larrin 's article on how to HT 52100.
Divorced Eutectoid Transformation Anneal

I’m not going to go too much into the details of the DET anneal, but it has been recommended by Dr. Verhoeven [2], who is known as a metallurgist who has done much for knifemakers. It is designed to be a relatively quick annealing treatment and to result in a fine carbide size. Devin first normalized the steel by holding at 1700°F for 20 minutes followed by an air cool. The steel was then given a grain refinement treatment by holding at 1460°F for 30 minutes followed by an air cool. The DET anneal was 1460°F for 30 minutes followed by a furnace cool to 1260°F (natural cooling of furnace), followed by an air cool. Dr. Verhoeven recommends 500°C per hour or slower for the furnace cool, the primary cooling rate used for his study was 370°C (665°F) per hour.

Here's the whole article:
https://knifesteelnerds.com/2019/05/13/how-to-heat-treat-52100/
 
You must log in or register to reply here.
Back
Top