5 hour soak with O1

Phillip Patton

Phillip Patton

Knifemaker / Craftsman / Service Provider
Joined
Jul 25, 2005
Messages
5,342
Howdy all,
I was following along with the "Logic of quenching oils" thread, and read where Kevin Cashen said that someone had soaked some 52100 for 5 hours without experiencing grain growth. Since I have the "gift of suspicion" :D and think that one shouldn't blindly believe whatever one hears, I decided to do an experiment and see for myself.

I forged out a piece of O1 about 5 inches long by 5/8ths wide by 1/4" thick, normalized it three times using descending heats, put it in a foil wrap envelope with some tissue paper and soaked it at 1510 for 5 hours. Cut the envelope open and dumped it in the quench oil. Ground a notch and broke it. Very fine grain. At least to the naked eye. I don't have a metallurgical lab (unfortunately) so I can't measure the grain size, but it looks pretty good to me.

It would seem from this experiment that grain growth really is more a product of heat than time. I'm going to try this with some other steel and see what happens. Are there any steels anybody would especially like to see this done with?

Edited to add, that I didn't think Kevin was lying about the 52100 soak, but it went against what I'd read up to that point. :D
 
Along these lines, during my first failed attempt to use my one-brick forge, I messed up a blade of O1.

Not knowing what I was doing, I left the maps gas on full blast. My blade got way overheated in less than 1.5 minutes. Not knowing any better, I quenched anyway. Later, while finishing the blade, it broke.

The grains are so large I can count them. Even with less than 1.5 minutes from room temp to quench, I got huge grains -- temp is what did it.

Hey, it was non-magnetic, what can I say :(
 
I have a feeling you'll get the same result with any steel you try it on. ;)

I just hooked up a thermocouple to my forge last night and I was surprised at how fast that puppy got up to high heat and I really had to turn it down to get it down to 1400.
 
I'm right with you guys on the soaking in a forge. My problem was not getting the blades hot enough. The even heat is probably the best investment I've made so far. But don't assume that it's at the temp you want. Mine was reading about 100 degrees higher than what it really was. As you can imagine, my knives weren't doing very well. :D Apparently it was a bad contact between the thermocoupler and the controller. I disconnected the thermocoupler and reconnected it and now it's right on. I'm still double checking occasionally with a spare thermocoupler and a voltmeter. Better safe than sorry.
 
Chant said:
Along these lines, during my first failed attempt to use my one-brick forge, I messed up a blade of O1.

Not knowing what I was doing, I left the maps gas on full blast. My blade got way overheated in less than 1.5 minutes. Not knowing any better, I quenched anyway. Later, while finishing the blade, it broke.

The grains are so large I can count them. Even with less than 1.5 minutes from room temp to quench, I got huge grains -- temp is what did it.

Hey, it was non-magnetic, what can I say :(
Click to expand...



This is why my forge hardened blades were doing so bad. I would quench as soon as they got non magnetic. I've since learned that most steels lose their magnetism at a much lower temp than you really want it. Plus I wasn't holding them at temp as long as I should have. I'm glad I didn't sell any of those. :D
 
Good for you, Phillip! :thumbup:

You taught yourself something of importance on several levels. After you come to see for yourself how this works on all your steels you will feel comfortable soaking them for the time necessary to get everything you need in solid solution to get the most out of your steel.
 
You folks don't know how pleased I am to see this thread. If Phillip had just taken my word for it he wouldn't really be any wiser, and the basic principles wouldn't have gotten such a concrete reinforcement. You can try it on other steels, if you like, but I am pretty sure the laws of physics will still apply to any ferrous alloy.

If you have any pieces of that slow roasted O1 that you would like to part with, I would love to slip it under the microscope. I know what happens with grains but my studies currently involve carbide dispersement, dissolution etc... I have no examples of O1 with that magnitude of soaking, and to be honest since you have already used your electricity, it will save on my bill;).

Also to give you and idea of what differences of control there can be in various methods of heating, you went to 1510F. most of my spec sheets show a MAXIMUM of 1500F for O1. So you have the control to let that stuff hover at the top limits for 5 hours, not bad:D and I moved from ovens to salts for even tighter control yet, I think we are all a little obsessive... nahhh:D
 
Phillip, I,too,am pleased to see a young and smart maker look at the metallurgy that is going on .Too many assume that if you get it red hot real fast....and cool it down real fast...it will get real hard.Unfortunately that is were they stop.Control of the heat is the heart of the blade.I hope a couple of dozen newer makers (and some old timers) order an oven this week,after reading your post.
 
I have an Even heat oven.

I always soak 52100 for between 4 and 6 mins at 1550....plus it takes a few minutes to get up to temp every time I open and close my door......My rule of thumb is that if I can see the grain, with my naked eye then its too big.
I use a loupe to check the grain and have recently stated using a microscope....My microscope is a bit over the top at the moment and I am not too sure what I am seeing...but it sure looks pretty in there:D .

Using my ht and soak times, grain has always been invisible to the eye.
 
Why should we make such a big deal about soaking? Because achieving a proper solid solution before quenching is important.

Here are some results of my most recent tests that you guys may find interesting. The O1 was soaked at the recommended maximum of 1500F. I used O1 because I want to know the stuff I see under the microscope is a results of improper soaking, not insufficient quenching.

As quenched hardness after soaking for:

2 minutes- HRC 62.6
4 minutes- HRC 64
6 minutes- HRC 64.7
8 minutes- HRC 65.2
10 minutes- HRC 65.3

In tempering, the less soak time the quicker the numbers dropped, with the 10 minute soak taking very high temperatures to soften. It is all about that carbon. I have done quite a few similar studies with 1095 that show the same results in higher tempering being necessary with longer soaks.

But one can clearly see what the maximum hardness may be if you do not have that ability to hold accurately at temp. Does this mean that a forge is no good at all, heck no, but some steels just won’t reach their full potential without accurate heats. How often have we heard from bladesmiths how much success they had with 1084? That steel is very simple with a carbon level that is close to the eutectoid, which simply means that it will go into full solution the quickest, so it is great for forge heat treating. You may have heard me say it before, but this is what I mean when I say the more complex the alloy the more complex the equipment to work it properly.

What I do find troubling in my studies, and the main reason why I am doing most of the microscopy, is despite much higher Rockwell numbers the spheroidal carbides are still similar from a 6 minute soak to a 12 minute soak. Phillip’s 5 hour stuff could answer a few questions about this.
 
Thanks Fitzo, Kevin and bladsmth, accolades from makers such as yourselves mean a lot to me. I might have to get some bigger hats. Just kidding. :D

Kevin, you're certainly welcome to the pieces. I'd be really interested to see what you come up with. That's interesting about needing higher tempering temps for longer soak times. I would think that would mean that the steel would be more wear resistant at the same hardness than if it had shorter soak times. Is that true?

bladsmth, maybe I should buy stock in Evenheat and Paragon, eh? :D
 
Kevin: You're collecting some good data there.

What I see when I look at it is that both temperature and time are critical.
Steel can be soaked for longer than most people expect without detriment, provided that the temperature it's being soaked at is at the middle or low end of the Austenitizing range.

And, steel requires sufficient time at temperature to get the alloying components with higher melting temperatures into solution. That is my suspicion for the 01 specimens you did having a higher rockwell after a longer soak-the Tungsten and Chromium need more time at that temperature to get into solution. Tungsten carbides and Chromium carbides are harder than Iron carbides, right?

What do you think?
 
Kevin: One other question: Is your hardness tester really accurate to 1/10 th of a point? If so, how many tests did you do per sample and what kind of tester is it.

THANKS
 
Typical specs for steel have a range of 2-3 points .Under laboratory conditions with one batch of steel you might get within 1 point. ...I don't know , off hand, about hardness of carbides but wear resistance for grinding can be very different . The reason that there are so many tool steels containing chromium is that the chrome carbide is easy to grind ,tungsten carbide is not! So we can list carbides in order of their wear resistance or grindability....As for heat treating we list the carbides by how strong the metal carbon bond is. Iron and chromium are the weakest bonds while Mo,W and V are the strongest bonds.
 
You are correct that there is a huge difference from the other carbide forming elements and simple cementite. Thank you for reminding me to clarify the fine increments on my HRC numbers, I thought it would raise questions and meant to explain that originally.

When I am testing this sort of thing I take multiple readings, one is just not good enough, and then average them out. Anything above the average range or below gets discarded and as some sort of error. If I cannot get within a point in my readings I will continue to take readings until a pattern develops, that I can glean some sort of data off from.

I know there are many who would read this thread and say "phooey, all you guys care about is rockwell numbers, there is a whole lot more to a good knife than rockwell numbers". To a very small degree there is a point here. Rockwell numbers are a good guidepost, but certainly not that whole story, the grain we are looking at is just as important. But at least rockwell numbers are standardized and translatable, without all kinds of skewable subjective concepts. Many of the improvised tests that I see for hardness/thoughness are hardley in the same category, any test that involves flexing of the metal in any way tells virtually nothing about the hardness of the blade unless it is taken far enough to either bend or fracture the steel, short of this it really only reflects the thickness that you ground it to, and you can see that with your own eyes.

If we want to convince the world we have found something new in the processing of steel we need to present data on that is translatable by the rest of the world, not just in our shop. We can ask mete if he has ever seen anybody in industry or a lab tug on a piece of steel clamped in a vice and then yell "eureka", somehow I doubt it. There is testing and there are parlor tricks. Testing generates impartial hard numbers that have a good chance to prove us wrong, parlor tricks are geared specifically to make our stuff look good for marketing. I still struggle with this every day, I have used a few parlor tricks over the years without realizing it, and am still trying to weed them out of my testing, I like to ask of the test, is there a chance this test could prove me wrong? Plus it is not cheap to test the way the big boys of industry do in you own shop. I have my impact tester to test the materials in general, but as soon as I get time I plan on starting on a tester that will actually tests blade cross sections for impact in a directly translatable way.

As you can see the testing never ends, this is mostly because when ever I find something that seems to be really big, and particularly against traditional wisdom, I consider it in error until I can absolutely prove otherwise, it tends to keep egg off from ones face.
 
I am not casting aspersions on your data. It seems to me that the longer soak times are allowing the formation of tungsten carbide, which is increasing your Rc hardness. What do you think?
Mete: I read your last post, but, it doesn't comment directly on this idea. What do you think?

RJ Martin
 
I have always viewed bend testing as a test of ductility. When qualifying welders, I bend the strips from the test coupon in a jig with a known radius and known thickness of the strip. Bending blades shows ductility, which is probably tricky to get when forging by hand and eye. It shows a certain level of skill in controlling the heat. It does seem to me that the hardened portion of the blade is in the center of the bent blade, in the area where the bending stress is the lowest. Perhaps a standardized test for bending blades could be used, with a known thickness and a known bend radius.

Kevin, since I failed at your "identify the micrograph" thread earlier, I'm going to try to redeem myself now. I'm going to guess that, at the lower soak times, the carbides arent disolving, and the elements in the carbides cant contribute to softening resistance as you increase the tempering temperature. Perhaps also, if the micrographs show no difference in carbide size and frequency, then the carbides are disolving, but the elements do not have time to uniformly distribute themselves. I read somewhere, perhaps Verhoeven, that the elements in low alloy steel can take as much as an hour to completely homogenize. I'm not sure what it is about the carbides that look the same.
 
You must log in or register to reply here.
Back
Top