Intro to Heat Treating

[anvil]

Steve, nearly all of the spec sheets list normalizing before anneal for most steels.

This is for the reason given by Buzzkill.

It's also why Cashen lists the same steps I list in all of his examples on his website.

Because you normalize at a slightly higher temp than annealing, grain growth is not affected by anneal. 

And I consider annealing to be a much needed step in the heat treat process. 

As I've said many times,,, you just can't lose by following the specs for the steel you are using. So, again,,, follow the specs, you are in good company,,, supported by Cashen.

I say nearly all because there are so many kinds of steels. I will say this step sequence works for all common steels used by knifemakers today.

I have never seen annealing not listed in the spec sheets. There are a few steels that the specs say do not normalize, and there are only a few steels that the specs list normalize more than once.  

 

[DJW]

Well thank you for all this information gentlemen when my head stops spining I will read again lol. I just remembered there is a reason I swung a hammer and and melted welding wire for a living and didn't become a nuclear scientist ! I'm reading as much as I can on here so not to ask dumb questions and if I do ask one it's because I couldn't find or just didn't understand it . Unluckily I threw my back out a few weeks ago so I have plenty of time to read . You guys are doing a great job and thanks again .

Dennis

[Donald Sawyers]

After reading how to temper in the oven, and then holding that temperature for 1 or 2 hrs, and then letting it cool, and then doing it again, I have to ask:  how does treating with various methods in the forge while watching the colors change work? It seems that it is heated for just a few minutes and then quenched.  Does the oven method create a better blade due to changes internally in the steel while the forge method just makes it less brittle?  

[Frosty]

Both methods make the stock less brittle. Tempering is reducing the piece's hardness to lessen brittleness or target specific properties, say a spring as opposed to a straight razor. 

Oven tempering produces a uniform hardness over the entire piece. And depending on the temperature controls a more precise finished hardness.

Typically tempering over the forge or with a torch produces a progressive temper even if the smith is aiming for a uniform temper because the heat isn't uniformly applied over the entire piece. 

Most knives aren't thick enough to require much soak time so using a torch or the forge will effect the full depth. However if the center is a little harder than the outside it's contained so it's not likely to crack or fail in use.

Bear in mind please, I'm not a bladesmith but have had cause to heat treat so I shouldn't be completely wrong. 

Frosty The Lucky.

[ThomasPowers]

You are less likely to foul up the tempering with an oven.  A skilled smith can do a differential temper using a forge or torch or tempering tongs.  Some folks do a snap temper in the oven and then go back and differential temper the blade using heat applied in a smaller area.

[anvil]

What they said. The spec sheet for your steel will give you all the info you need to do a proper heat treat. 

Sometimes people don't understand the spec sheets. Like frosty said, it depends on the cross section. Concerning soak times, the data is based on a 1" dia piece of steel, I think 6" long.A set length anyway. If your oven is preheated to a set temp, then a 1 hour soak means it takes that long to get a constant heat thru the piece. If you have a 1/2" crossection, it only takes 1\2 hour and 15 minutes for a 1\4" crossection. If you dont know this, then an hour works,,,  about the same as letting it soak for a week. In this case, anything over 15 minutes for a knife is not needed. 

And as far as which is better, well its the opposite of what you stated. An oven actually limits what you can do to a blade when heat treating. Like was said above, your whole knife will have the same temper. What does this mean? If your steel holds the hardest edge at a dark straw and your whole knife is brought to this color (colors equate to specific temps all things considered) it will break easily in use. To prevent this, you have to soften it more to about a spring temp. So you lose a little on the edge holding, but your knife wont break. A compromise. No compromise with a forge. With a differential temper, you can get a soft spine to absorb shock, a spring temp in the center and thru the transition from blade to tang, and a nice 1\4"-3\8" wide straw for the cutting edge. For most tool work, an oven is just a convenience. 

[Donald Sawyers]

Thanks Anvil,  that cleared up a lot of confusion for me.  

[Bud in PA]

Are there any thermometers that we can use to determine the critical temperature of steel? Something that the average smith can afford.

[TWISTEDWILLOW]

I think John did a post on handheld infrared thermometers here recently but I don’t remember where I saw it,

[JHCC]

 

[Latticino]

I purchased one of those from Amazon as well.  Around $60, if I recall correctly.  Seems fairly accurate based on comparisons with thermocouples in a heat treat oven, but have read other postings from "authorities" that indicated there can be fairly large discrepancies based on the size of the cone and emissivity (note that the one he is commenting on is only rated to 1,500 deg. F, not the same as the one John and I have):

This from Jerrod Miller (a metallurgist):  I have yet to find one I trust for what I would call a reasonable amount of money.  Two big things that a lot of people do not realize with IR thermometers is the size of the cone for the reader (close to the gun reads a small area, farther away a bigger area) and emissivity constants (a material dependent value).  When we get to the temperatures we care about in smithing, emissivity constants become important and you will need to calibrate and adjust it for the material.  The cone for the one you linked has a 12:1 ratio, so at 12 inches your diameter being measured is 1 inch.  So unless you can be sure to be measuring just your steel, you will get contaminated readings (measuring the forge wall with the wrong emissivity constant for it). 

[anvil]

Thanks for the quote. I did not know the physical limitations, but am aware of the basic coefficient of error in most measuring tools that, like he implies, decreases slowly as costs quickly rise. 

[JHCC]

21 hours ago, Latticino said:

The cone for the one you linked has a 12:1 ratio, so at 12 inches your diameter being measured is 1 inch.

Would that mean that the measured diameter would be smaller if the thermometer were closer to the object being measured? Would that still be accurate?

The thermometer I got has a 30:1 ratio, and it has two lasers that converge 30" out, to show when you have it at the proper distance. Would I get an incorrect reading if I were holding it at, say 15"

[Latticino]

As I understand it the area being tested is proportional to the distance the test is at, and the temperature is averaged within that area.  Apparently the mechanism used for determining the temperature has some relation to the emissivity of the material in question and can be confused by metering multiple materials simultaneously (and Jarrod indicated that was worse at elevated temperatures, which doesn't make sense to me - but I'm not a physicist).  Bottom line is that closer sensing is probably fine up to a point (probably related to the size of the optical sensor being used).  He may be concerned with accuracy within +/-0.5% , while I don't need quite that level of accuracy at 1,500 degrees (+/- 1 or 2% is probably acceptable for the heat treatment I do).

[natkova]

Hi i got confused when i  quench steel and i want to see colors it get dirty.

Do you guys grind shiny before going to fire? How to do it so it dont oxidize , or get dirty.

I had luck couple of time set it in fire (steel) than as it is non magnetic i quenched it, then set it in anvil and used a litle bit file.

But i saw blacksmith using angle grinder first than quenching.

I heat treated center punch chisel like too. Its realy hard to  see tempering color. 

[Latticino]

Unfortunately your question is unclear.  It isn't obvious whether you are looking to determine temperature before hardening, or during tempering.

For Hardening:  Temperature needing to be reached is just when the particular steel enters into the austentitic range.  Best way to tell this is looking for decalescence.  

I wouldn't use tempering colors to gauge temperature for tempering after the hardening cycle on blades.  Works OK for chisels, but I want to be more accurate for blades.  I would strongly recommend you temper in an electric oven.  You only need to go somewhere between 350 and 550 deg. F (depending on the hardness desired).

[anvil]

"  Do you guys grind shiny before going to fire." When I'm running colors, your steel should be shiny and have no mill scale on it. I do this with a file or sand paper after removing it from the fire. And yes, the steel is hot, so use a "hot" file or sandpaper wrapped around a block. 

[natkova]

20 hours ago, Latticino said:

Unfortunately your question is unclear.  It isn't obvious whether you are looking to determine temperature before hardening, or during tempering.

For Hardening:  Temperature needing to be reached is just when the particular steel enters into the austentitic range.  Best way to tell this is looking for decalescence.  

I wouldn't use tempering colors to gauge temperature for tempering after the hardening cycle on blades.  Works OK for chisels, but I want to be more accurate for blades.  I would strongly recommend you temper in an electric oven.  You only need to go somewhere between 350 and 550 deg. F (depending on the hardness desired).

I know that for hardening and you need to check with magnet its easier.
But with tempering its hard to find that straw blue peacock or any color, metal is too dirty.

 

1 hour ago, anvil said:

"  Do you guys grind shiny before going to fire." When I'm running colors, your steel should be shiny and have no mill scale on it. I do this with a file or sand paper after removing it from the fire. And yes, the steel is hot, so use a "hot" file or sandpaper wrapped around a block. 

How much time you have before tempering collor apear , i allways think i will loose time while i file it to get that color.

I tought it could be done by grinding then quenching then after you quench you can see. But you cant scale form on it and there is no that white shiny space.

 

Dont get me wrong i know for what iam looking for but  sometimes its hard to predict when it will appear.

That's my bigg issue with heat treat.

 

 

[JHCC]

The tempering colors come from the steel oxidizing as the heat increases. The speed with which the tempering colors appear varies depending on the size of the workpiece and on how much heat is being transferred to it, either from an external source (a torch or a heated block of metal) or internally (from an unquenched part of the workpiece -- i.e., "autotempering"). A thin workpiece will show colors very quickly, while a thick one will be slower. A hotter heat source will make the colors run faster, while a cooler one will be slower. It really is a matter of paying attention and learning from experience.

When autotempering (e.g., making a punch), I remove the tool from the heat at the proper temperature, quench the working end, and then shine up the surface with an abrasive. (A piece of grinding belt or a broken grindstone works well.) This removes all oxidation from the surface and allows it to show the tempering colors as it re-oxidizes.

[natkova]

When you shine up one side does it mean the other side have same color?

Lets say  you shine up 3/4 on one side  did color traveled same thicknes trough whole piece.

 

[Latticino]

I reiterate, it is a bad idea to depend on temper colors to properly temper knives IMHO.  These colors can be affected by a lot of different things (traces of oil on the surface, ambient light, fingerprints...) and are unreliable.  It isn't as big an issue for struck tools where a hardness range is acceptable (and I use a very similar technique to JHCC), but ideally not to be used for blades. 

For blades I would temper in an oven.  If I had to torch temper (or use tempering tongs) I would completely abrasively clean off the thinnest sections of each portion of the blade and carefully heat from the spine down to the edge.  You should let the blade go down to room temperature to maximize martensitic transformation anyway, so no reason to not fully clean the edges.  Definitely safer if you give it a quick "snap temper" first if possible.  

Note: if you are torch tempering, you need to work your process so you heat both sides of the blade equally.  Not a big deal for a thin kitchen knife, but significant for an axe...

[JHCC]

24 minutes ago, natkova said:

When you shine up one side does it mean the other side have same color?

Because the temper colors are caused by the surface of the steel oxidizing at certain temperatures, then No. If you do not clean up the other side, it will still be oxidized from the fire and therefore will not have the same color. However, that does not mean that it has not reached the same temperature; it's just that you won't be able to tell.

[anvil]

JHCC's method is mine as well. I almost always do a Differential temper/reserve heat method as he described. This means the heat runs from the back or spine twards the cooled and hardened working end that was quickly buff'd as JHCC said above. My belief on the colors is a bit different. I believe the scale maskes the colors and makes them harder to see. They are still there and can be seen, but its difficult to see them and time is critical. . So, for me, If I polish only one side, or edge, assuming my heating to critical was done properly, and my quench to harden was done properly, And I'm working in my familiar environment especially having a constant light then the colors I see on one side represent the true color throughout my piece and thus, the temp is the same throughout. When doing a blade the process is the same with the addition of a way to supply water to specific places where the color run is too fast and a way to add heat to the spine where the color run is too slow. 

Lol, actually I think JHCC and I are saying the same thing,,,

[natkova]

Well i red and its true when you want to quench  room need to be darker so you can see that glowing temperatur.
When you want to quench.

But when temper i would say it need to be light so you can see that white straw transforming to other colors. And i gues you "catch " that color and cool material.

 

[anvil]

No, what happens is that if the temp doesn't change, the color you see will be different in say direct sunlight than that same temp in the light of your shop. So, in your shop, if it has a constant light from morning until evening, the color you see will always represent the same temp.