[Joshua Snead]

This is my first time performing a heat treat on my own, so, before quenching the blade it's taken so long to make, I practiced on a piece of 1084 from the same stock. I heated the piece just past the point where it lost magnetism and quenched it in canola oil. I tested it with a file and it clearly had not hardened so I tried again, this time based on color. When it looked about cherry (presumably around 1500^oF) I quenched it again, but again it wouldn't harden. What is the best way to determine the right time to quench the steel? I am using a simple, atmospheric gas forge that is open on both ends. The forge produces a lot of scale and I did not remove it before quenching. Also, the test piece was about 3/16" in thickness. Obviously, I don't have much experience with heat treatment, but I know I must be doing something wrong. I'm not sure if the issue is the quenchant, steel temperature, scale, or something else entirely. Any advice would be appreciated
Thanks,
~Josh

[Ed Clarke]

Josh,
Hi there . I have used canola oil and 1084 for a while. A couple of questions first. Where did you get that 1084 and are you sure it's 1084? What temperature was the quench oil? Taking the steel up to just past non-magnetic and then quenching immediately should work. Don't forget to keep the blade moving gently(a figure eight is effective). The oil needs to be 100-120 degrees F (no hotter). If the oil is too cold, you will create a vapor jacket around the blade which will actually insulate it enough to keep it from cooling fast enough to harden. The hot oil and keeping the blade moving (not fast) will prevent this from happening. That's my 2 cents from my experience...maybe a Master Smith will check in with more/better info. Hope it helps!

[BrionTomberlin]

Josh you may want to do a little bit of grinding on that blade and see what happens then with the file test. You may have a layer of decarburization. Once you get past that you probably have a hard blade.

Like Ed states, canola oil at 100 - 120 should work fine and move the blade gently in a back and forth motion. 1084 should harden fine in this quenchant.

Try the grinding, because my bet is decarb. Let us know.

Brion

[Kevin R. Cashen]

Josh, what is the method you are using to determine hardness? This is not to question the method but to help in narrowing the possible problems based upon what symptoms are being read.

I have been working on several writings that include basic trouble shooting of heat treatments as it seems to be one of the most common source of questions in knifemaking, so do not feel bad you are not alone at all. In trouble shooting the hardening operation I have found several most common places to start looking. First you start with the most critical and most basic, in order for steel to be hardened you need to put its carbon into solution with heat and then trap it there with cooling so-

Was it hot enough?

Did it keep its carbon?

Did it cool fast enough?

I list these in the order of probability of being a problem. Just to non-magnetic can be tricky since that state occurs at 1414F for iron while most steel will need around 50 degrees more to start getting carbon fully into solution. Now this changes like a seesaw with time at one end and temperature on the other. The more temperature you have the less time it takes, the less temperature you have the more time it takes, and alloying will add significantly to time, but we are talking about 1084 so we can still keep it simple and only worry about carbon. With 1084 you may want around 75 degrees higher than non-magnetic if you are not holding it for any amount of time, often in the process of bringing it up in the forge you get there and lose it several times before seeing a nice even color, this needs to be considered in the estimate.

In addition to the magnet I would use a scrap bar of your steel to practice recognizing decalescence in the heating. This is a distinct dimming of the steels glow when it undergoes the transformation that puts the carbon into solution. It is best observed in total darkness until you get the hang of it. Heat a bar from one end only and watch the heat creep up the steel. You will notice a bright orangish-yellow band of heat at the fore front followed by a duller red band; that duller red band is decalescence. It takes added energy for the steel to make the crystalline changes necessary for solution and that dull band is where the extra energy is being pulled back into the steel instead of being sent to your eyes as light. Now on the other side of that dull band there will be another bright band, this is steel that has passed fully through decalescence and is in solution, heating beyond temperature is overkill.

The next question is- did you get it more than hot enough, long enough, and end up losing the carbon necessary to harden? As soon as you start going incandescent the carbon in the steel will begin to combine with oxygen and leave the steel to join the atmosphere. He hotter you go the more the carbon can move and be available to the atmosphere. The more oxygen in your atmosphere, or sometimes the moisture content as well, the more carbon can be lost. This is what we commonly call “decarb”, and is much more common than any of us care to admit. Almost any time you heat the steel without shielding from the atmosphere you will get some degree of decarb, often this is only patchy and a few thousandths thick at the surface, but it can easily get bad enough to give the appearance of unhardened steel. Does you steel have a marbled, raised patchy look to its surface that looks like the reverse of somebody dripping acid on it? Check it with a file, does it bite really well but after several strokes start to skate better and better? Well, then you have a skin of decarb. The good news is that you can probably remove it to reveal the hardened knife underneath by grinding about 3-5 thousandths off the outside. The real problem with decarb is that it gets worse with time at temp, so without protecting the blade, proper soak times can increase the threat.

Although most people suspect the quench first in these situations, in blade size cross sections it is actually the least likely of these first three areas to be the problem in under-hardening. Under 3/16” most oils will result in some hardening, and at the edge, where we are often closer to 1/16” we should get some results. So long as you are under the surface of the quenchant before recalescence (the opposite reaction on cooling) you should get some hardening. I would like to point out that this is very misunderstood by many people. It is said that you have less than .75 seconds to cool 1084 below 1000F to fully harden, this DOES NOT mean a frantic and dangerous dash from the fire to the quench tank, it means that once you begin the fast cooling that takes it from 1200F to 1000F the QUENCHANT needs to be that fast; please take your calm an safe time handling 1500F steel in the open air.

Here practice in observing recalescence can be even more of an education. Heat a bar of steel that same as before but this time in total darkness observe the bright band that will flash from the darkening steel as it changes back from the in-solution state. You will be astonished at how long it takes for this to occur after you remove it from the heat, and while you should leave nothing to chance and keep your quench tank right beside your forge/oven, you will be much more at ease about the 1 second you lose from heat to quench.

I have found most problems in hardening can be traced to these three areas. When they cannot, then we need to start widening our search. Just like after troubleshooting an electronic for an hour or two you finally get desperate enough to check if it is plugged in, you eventually cannot overlook the possibility that the steel you have may not be what you think it is, materials are often mislabeled. Although you have clearly and correctly indicated 1084 in this case Josh, for many other who could read this, I must take the opportunity to implore folks to always say what the steel is when asking others to help troubleshoot. If it is 1095, include that information in the details, if it is old saw blade or old spring, please, please, say it is old saw blade or spring, and do not label it L6 or 5160 because that is what it should be. I have spent many unnecessary hours trying to help folks based on a chemistry they did not have.

If the mystery remains unsolved, then you can gather more data by supplemental testing. Take the same steel and overheat a sample before quenching in your oil. Did it harden? Take another sample and heat it carefully to where you think you should and quench it in water. Did it harden? Check both with a file, a Rockwell tester, if you can get access to one, and then break them if you can. How did each behave with these tests? You will learn a lot from this and the 1.5” samples will be a lot less costly in time and material than a blade.

I wish you luck

[Joshua Snead]

Thanks everyone for the advice! All this info is very helpful, and especially thanks Kevin for putting all that info in an way that can be easily understood. I saw a bladesmith demonstrate decalescence before, but never understood what was actually going on until reading your post. As for my heat treat, it seems I may have several things going on, but first to answer questions:

• I bought the steel from USA Knifemaker Supply so it should be 1084.
• The oil was hot during the quench, but I do not know the exact temperature.
• To test the hardness I cut along the edge of the bar stock with a file; I don't have access to a Rockwell tester.

I did not move the piece at all during the quench so that's at least part of my problem. I'm fairly certain the steel was not hot enough during the second attempt. The first attempt may have acttually hardened in which case decarb would probably be the issue. I will try again when I have some free time and let you know how it goes.
Thanks again,
~Josh

[Shawn McIntyre]

G'day Josh

The first thing I noticed about your post was when you said "another piece of the same stock", not a blade. If you have a close look at Kevin's comments as regards cross section and cooling rates you may have your answer.(have a close look at all of it, really good stuff there, I am just putting my money on cross section)) I have a tank of vegetable oil in the forge area for quenching tools and oil finishing black iron work. I have done quick demos for people and have quenched flat stock 1084 in the tank without preheating the oil. The steel did not fully harden, not even close in fact. The material acted more as though it was case hardened. You could put it in the vise and smack it with a hammer and not break it although it resisted bending, really resisted. The same steel, in knife cross section, quenched in warmed vegetable oil will go hard enough to skate a file and snap like glass in the vise. I would also recommend always leaving an allowance for decarb. People often allow material for grinding from the bevels of the blade, this is natural as you want to leave the edge thick enough to avoid warping. But they sometimes fail to leave material to remove from the profile of the blade. Decarb comes at you from all directions and as the edge is thinnest it will heat faster than the rest of the blade. In an unprotected forge environment you will more than likely lose carbon in the edge and point but the spine and ricasso area will be fine. Exactly the opposite of what we would want! Good luck and let us know how you progress.
Cheers
Shawn

[Joshua Snead]

sorry, double post

This post has been edited by Joshua Snead: 06 February 2012 - 04:33 PM

[Ed Clarke]

Brion and Kevin,

Thanks...I wasn't even thinking decarburization, but I should have been. You have opened my mind...

[Joshua Snead]

I am fairly certain now that the issue was decarburization. I performed a bend test on the same piece of steel I mentioned before, without doing anything else to it, and it snapped in two fairly easily. The center looks evenly hardened and skates a file as well. Thanks for all the help! Now to figure out tempering...
~Josh

[BrionTomberlin]

Joshua, on 1084 I do the standard 2X2 temper. I use 400 degrees for two hours, let it cool completely and do it again for another two hours. I will check the edge then and if needed raise it to 425 degrees for another two hours. The 400 degree should give you between 58-60 on the rockwell scale.
Brion

[Joshua Snead]

Thanks Brion,
I finally got around to hardening my latest blade, and it went really well! I was amazed it didn't warp at all. I put the blade through the 2 two-hour tempering cycles at 400 degrees as advised. It came out a dark straw color. Now I am wondering, how exactly do you "check the edge" and what do you look for to make sure it's reached the right hardness? Should the edge still skate a file after tempering?
~Josh

[Lin Rhea]

Light pressure with a new file will skate, hard pressure will cut, but you have to have a heavy hand or it will skip on you. When it acts like that, your good to go.

[BrionTomberlin]

Joshua, Lin beat me to it. I will say the same thing. I will also sharpen and test the blade on some hard oak 2X2's and maybe a brass rod test for deflection of the edge. THen I will check for chips. No chips and I am good, small micro chips, raise the tempering temperature a little bit and try again.
Brion

[Joshua Snead]

Thanks for all the help! I think the temper is about right now. Haven't sharpened it yet though. I will post some pics once I finish the blade.
~Josh

[BrionTomberlin]

You are welcome Joshua. Let us know how it goes.
Brion

[Richard Marchand]

Lin Rhea, on 21 February 2012 - 08:19 AM, said:

Light pressure with a new file will skate, hard pressure will cut, but you have to have a heavy hand or it will skip on you. When it acts like that, your good to go.

Keep in mind that everytime you perform this test, the file gets more dull.


I realise this may affect the final hardness(some say it may increase the speed of the quench) but I use an electric kiln for HT and experienced a lot of decarb so my solution was to to coat the blades in clay slip. Just a very thin layer. It made a huge difference. I get very little decarb. The blades are easier to finish and I don't have to go digging for hardened steel anymore. I couldn't tell you if I have gained or lossed a few HRC points with the clay slip but I am more than happy with the results.

[Kevin R. Cashen]

Richard Marchand, on 10 March 2012 - 10:44 AM, said:

Keep in mind that everytime you perform this test, the file gets more dull...

Controls are very important as a baseline in any test. One thing I always show students about the file is to use two parts of the knife. The tang is often softer or not hardened at all and I always teach my students to first cut the tang with the file they are using and then test the edge so that they can feel the distinct contrast between the two. It is kind of like calibrating your file, regardless of its condition. This methods is particularly useful when the knives have been tempered and will cut a little and one needs to know how much is cutting a little and how much is dead soft.

By the way Rick, welcome to the forum, it is good to see you made it here. For researching bladesmithing stuff you will find this forum like a library or attentive lecture hall as opposed to trying to study in a rowdy saloon on a Friday night.

[Richard Marchand]

Kevin R. Cashen, on 10 March 2012 - 11:04 AM, said:

By the way Rick, welcome to the forum, it is good to see you made it here. For researching bladesmithing stuff you will find this forum like a library or attentive lecture hall as opposed to trying to study in a rowdy saloon on a Friday night.

Yeah... I noticed that straight away. You may laugh but I get that erie feeling like something bad is going to happen. You know, like when dog is rescued from an abusive kennel and put into a new nurturing home... it still cowers when you reach out to pet it.

[Dan Petersen]

There is a lot of good information in this forum and most of it is focused on decarb which is important. Some of it is asking you to train your eye to see the color shifts as the steel absorbs or loses energy (heat) to transition from one phase to another. That is a skill you need to acquire. I would suggest another learning exercise you could perform to learn more about what is happening with your steel. Cut or draw out a piece of your steel so that is is roughly about a quarter to a third of an inch in diameter and about a foot long. Place this in your forge so that one end gets quite a bit hotter than non magnetic and the other end close to your tongs is quite a bit below the temperature you would quench from. Try to get the middle of this foot long piece of steel right at the temperature you think is the correct quenching temperature. So it is very hot on one end and much cooler on the other. Quench the whole piece just like you would a blade. Now let it cool and then do the following (do not temper). Start breaking off 1 inch or so pieces from the hot end to the cool end. You should be able to do this easily on your anvil with a hammer. Look closely at the grain structure of the broken pieces. What you should experience is the following: It should be easier to break the pieces of the end that got very hot and the grain structure will look large. If you got the middle right the grain size should be getting really small and it might be just a bit harder to break it. Towards the cool end it should get noticeably hard to break the steel and you might find it just does not break with a hammer. When you get to that point, stick it in a vise and bend it back and forth until it breaks. The grain size should be fine and it should bend quite a ways before it breaks.
The exercise will do a couple of things. It gets you to see the changes in grain of the steel based on temperature and allows you to get feedback on whether the area you thought was right on the money for quenching the steel, really is. Also, if you get large grain in the steel where bending was required (colder end of the piece), then you probably needed to do a better job of normalizing the steel before you try hardening it.

Dan P.

[Joshua Snead]

Thanks Dan,
I performed that test with an old file and used it as a demonstration piece for a Chemistry presentation. It's really cool seeing the different grain sizes. After quenching a blade, do you always want the grains to be as large as possible?

Unfortunately, I have some bad news. I have been working on this blade since the beginning of last summer, and I really liked the shape, but something happened. As I was hammering the guard onto the tang to get it nice and flush, I looked at the blade and, to my horror, it had snapped in two! It's upsetting to have that happen after so many hours of work, but at least I've learned a lot (including to always make a backup). Here's a pic of the broken blade:



Thinking back I realized that I forgot to draw the tang. (Did I mention I did a full-blade quench?) Here is a picture of the broken edges, sorry if it's difficult to see:



I don't know why the spine is a darker brownish-straw color. Could that be a result of poor or not enough tempering?

Thanks & Happy Easter,
~Josh