This past weekend I made my first attempt making a knife using O-1 tool steel. I started out with flat stock and using a Whisper Momma propane forge got it into the shape I wanted. I was not ready to oil quench it quite yet, so I let the blade air cool. After 30-45 minutes, it was cool enough for me to handle and, wanting to see how it looked cleaned up a bit, started to clean it with an angle grinder with a wire brush attachment. I had the tang of the knife in an vise and was applying some pressure when the knife snapped.





Before doing anything work to the steel upon finishing, should I have tempered it? Besides breaking it, is there something I should have done? A friend of mine made a blade at the same time and is worried to do anything else to it.

Any thoughts?

Remember that most of the quenching information is based on a 1" thick section. Knife thicknesses often will quench one or two steps differently. So it's possible that your piece actually hardened and then you were working untempered and so broke.

How was it being supported during "Air Cooling"? What was the ambient temperature of your shop? Still air or was there a breeze?

Why not test your friend's (and your) blade with a file to see if it has hardened?

Did you do much hammering on the steel below a bright red heat? If you hammer on steel much after it has cooled you can caused stress. I have had a tool of two break from hammer to cold. It can be tempting to just hammer those few more times after it has gone to a black heat.

The grain at the break seems coarse to me. Perhaps it was overheated and did not normalize. As Thomas said, thin sections will get quite hard with a chill - in fact, if you laid it on a heat sink like a cold anvil or table, it would equate to a very fast quench.

In addition to wot was written above,, you are not the first person to find out that O-1 air hardens.I believe it is easier to start with a simple steel like 1084
If you try it you will I bet be alot happier.

I appreciate the responses, they were all helpful.

The blade was being supported on a metal dog cage in a garage. Will not make that mistake again.

There was some hammering below red after a couple of heats. I may have done it on the last heat just to straighten it a little bit before I called it finished.

.

Why not test your friend's (and your) blade with a file to see if it has hardened?

Sorry to be the new guy here, but I was reading this post and was curious if you could explain to me what exactly the file test would be?

I have an old "Bethlehem Tool and Die Steel Manual" that says this about O1. "Normalizing. Do not normalize." I have read elsewhere that O1 will harden in air, but that internally, it will be unstable. When finished with forging, O1 should be annealed.

I got into it on another forum one time when I told the "bladesmiths" that you couldn't thermal cycle O1, which they were trying their darndest to do. They jumped all over me, telling me that I didn't know squat. They were going to repeatedly air cool it. What did I know? I went to Whatsamatta' U?

I would have to disagree with you, both from experience and current literature.

The heat treaters guide 2nd edition says this......
The O series steels are generally normalised to produce a more uniformly refined grain structure, especially after forging or previous heating to temperatures much higher than the recommended austenising temperature.

specifically for O1
Normalising.
heat to 870C , after uniform through heating work is cooled from temperature in still air.

from personal experience:-
O1 grows huge grain at forging temperatures and responds very well to normalising showing a dramatic change visable to the bare eye ,with sucessive normalisations , the first being most dramatic.
In this large grain state I have broken O1 cold by accident by hammering further along the bar, its very brittle . Its a combination of its ability to grow grain and air hardening that can cause problems with O1 .
it can be a great steel though , but not my number one choice for forging.

The file test is a common test to check if something has hardened: take the piece and try to file on it with a new sharp file. If the file slides like it's on glass the piece is *hard* (suitable for knives); if it will bite with quite a bit of pressure, it's hard (maybe suitable for axes), if it files it didn't harden with that heat treat.

WARNING new smiths will often have a substantial decarb layer on their pieces that will give a false indication of being soft but if you continue you may get to where the file will slide like glass under the decarb layer. This also makes a difference with fire steels that should have a possible decarb layer removed after hardening and before use.

could you use an oven to normalize O1 steel by setting the oven at a high temp, put the steel in, turn the oven off and let it cool down on its own? just a curious question since ive never worked with O1

The temps used for normalizing are above 1000 deg F so a kitchen oven will not work; a Heat treat oven will. Note too that leaving steel in the oven as it cools down is generally an anneal not a normalize. Normalize is usually done in still air.

Heavy sizes should be left to cool in vermiculite or in other insulating Media.

Annealing 1425-1475° F Cool slowly in the furnace at a maximum rate of 35° F per hour to approximately 1110° F. Protect against surface decarburization using controlled atmosphere or by pack-annealing.

Stress Relieving 1200-1250° F Stress relieve after rough machining

Preheating 1200-1250° F Preheat time in furnace is ¾ Hr. per inch of thickness. Heating up to temperature as slow as possible.

Hardening 1450-1500° F In oil Remove all forging or rolling skin prior to hardening. Temper immediately after hardening. Allow to cool to 125-150° F before tempering.

Tempering Complete equalization of temperature throughout the tools is essential for good results. Recommended temperature is 350-500° F.

mod: cleaned up the sloppy mess of a post

AS I read the last post it seems as if the information is either out of sequence or may cause a mis understanding of the procees of heat treatment. Up front it tells us temp ranges for forging and then says lime or ashes. Then it says to anneal by heating to a specific temp and leaving in furnace for a controlled cool down rate of so manyh degrees an hour. If you read this back either as I have kinda quoted it or in above note, I think it can be less confusing. For all of the new folks that read this thread I feel that is really important.

Sorry for any misunderstanding rich here is the link so you can see the chart did not copy well hope this simplifies it.

http://www.diehlsteel.com/products/o-1.html

:)

To quote a bladesmith friend of mine (the mighty Col KC) , O1 will harden if you so much as fart near it....

I have accidently air hardened it and snapped blades, and rested it on the anvil which has quenched it, snapped blades... Although it can make mighty blades I have not had much luck with it. Ive also had some commedy tear aparts patternwelding with O1.

The UK's no.1 competition cutter gave some of the best a run for their money at Bladesports Intl. at Atlanta with an O1 comp knife, against all the 'super steels'

The grain in the broken blade is massive. Normalise a time or two before the heat to quench. There are more forgiving steels out there!

What would be the correct order in forging a knife out of this material? Would you forge, anneal, normalize, harden and then temper? Haven't had any experience with knifemaking yet and would like to know the correct steps before I proceed. Thanks.

REd there is a heat treat sticky on here that will likely help you alot and welcome to the site!

Annealing is generally done if you will be drilling or filing on the blade to shape it before heat treat. If not I'd skip the anneal and do the normalization x 3.

Save for: if you have to stop working on a blade say overnight (or longer) I'd do an anneal on it so I'd know it was *safe* until I could get back to it...

Thanks for the replies, I'll have to check out that heat treat sticky.

FWIW, a traditional vermiculite or wood ash anneal from well above critical, with steels like O1 and 1095 (in the .85%+ carbon range), can cause problems. Given sufficient time as austenite, the carbon will tend to clump up into big, nasty carbides that can make drilling and filing a real bear. Go to any bladesmithing site -- probably including this one -- and search the archives, and you'll find plenty of guys wondering why they're ruining drill bits on "annealed" 1095, O1, etc. It's because they used the old-fashioned, slow-cool anneal from well above critical, and formed big, highly wear-resistant carbides in the steel. (It's probably worse in 1095, because there are fewer alloying elements in the steel to keep the carbon from moving around, compared to O1 and other tool steels. Regardless, big carbides also aren't necessarily good for the blade if you don't get rid of them before or during the final heat treatment, which isn't the simplest thing to do.) If you notice the recommended annealing temp for O1 at the link Frank provided, above, it tops out just barely inside the recommended hardening range. There's a reason for that.

One of my earliest blades was made from O1, and despite repeated tempering cycles at increasing temperatures, I had a really difficult time sharpening it. Not that it wouldn't get sharp; it just took an incredibly long time -- I'm talking many hours -- with my coarsest stone to get it to start coming around. (And no, the edge I started with wasn't all that thick.) I now suspect that was at least partly the result of bad heat treating practice that left the steel chock full of great big carbides.

What I'm saying, essentially, is to be very careful when you're annealing high carbon steels, or you could end up making matters worse. A better solution than the traditional anneal is a sub-critical anneal or spheroidizing anneal, but that's not so easy with very simple equipment. http://www.steeluniv...g/3630-0050.htm

As far as I have been able to determine, normalizing is a function of time at temperature. That is you heat the metal to the specific temperature specified for the steel and then let it soak for a period of time depending on the mass of the piece. Then the piece is allowed to cool in still air. Nowhere have I been able to find a documentable reference that says there is anything to be gained by repetitions of the process. One of the sources I have queried is a professor of metalurgy at the South Dakota School of Mines and Technology. Could someone please point me in the direction of a publication on metalurgy that documents the benefits of repeated normalizing?

Didn't someone post pictures of the grain reduction over several normalization cycles on this very site?

Not a publication on metallurgy per se but experimental data tends to trump theoretical in my experience.

Yes, I posted that.

http://paleoplanet69529.yuku.com/topic/47099/Normalization-Grain-Size-Control-Experiment-----normalize

Metallurgy does explain how and why it works, but it's a little more complicated than I think I can adequately explain. (By which I mean that I don't completely understand it myself.) It has to do with nucleation of new austenite grains along the old austenite grain boundaries. (And austenite grain size determines the grain size of the room temperature structures like pearlite and martensite.) So if you form a bunch of smaller grains by one normalizing cycle, then austenitize again, the new grains will begin to form along the boundaries of the old grains -- and they'll necessarily be smaller than the previous set, at least until grain growth kicks in. But there is some practical limit to how far you can go with this -- or how far you'd want to go, since the finer the grain size, the less hardenable the steel. It is possible to refine grain size to the point that the steel will not harden properly, at least in some steels.