Drawing the spine/variable temper

[Justin King]

I have tried this a few times with a MAPP gas torch and am not sure that it is hot enough. I can only heat in tiny sections, get spotty results and am not getting the color to run as close to the tip as I would like, and the whole process seems to take longer than I think it should. Do most of you guys who do this use an oxy/acetylene torch for this?
I have heard of tempering tongs and am considering making a set to see if this makes the process any quicker or more consitent, but I have never seen any and although I have an idea what is needed, it would not hurt to get some advice and maybe see a pair first.
I appreciate any input.

 

[mete]

You should quench and temper the whole blade first. For the spinal temper if you use O/A use a rosebud tip and keep the flame moving fairly slowly back and forth along the spine. The edge and tip should be kept under water . It's difficult to tell the temperature from color and tempering is a temperature/time situation so the results are questionable.

 

[pike3e]

I have done it with both types of torches. It is faster with OA but it is not necessary. Keep the torch moving and keep the blade a little out past the inner cone of the flame. One other thing I have done is to put the blade in a pan of water so that only the edge is under the water. You can heat the spine more without worry of messing up the temper of the edge.

 

[Justin King]

I should clarify : I am trying to draw the spine on through hardened blades that are already oven-tempered. I am immersing the edge in water and this seems to be creating a large enough heat sink that I am having trouble getting the torch to heat the spine evenly. I am using a small Mapp gas torch like plumbers use, I do not have an O/A torch. The results just seem to be spotty and I am not sure I have the right tools/technique.
It also just seems to take forever. I figure the tongs will allow me to heat larger areas at a time and provide more positive heat transfer. Thanks for the replies at any rate, this is a subject I have not found much specific info on.

 

[Ed Caffrey]

My thoughts on this are as follows:

I think that fully hardening a blade, then trying to draw the spine back is taking the "hard road", and making things difficult on yourself.

I personally think you can achieve much more consistent and controllable results by differentially hardening a blade. To explain: I've had a fairly good number of folks come to me for their ABS JS and MS testing. MOST of the failures came from folks who fully hardened, then tried to "soft-back draw" the spines of their test blade. What you have to remember is that time AND temp have everything to do with it. Specific temps, for specific times is what creates the changes in the steel's matrix. Soft back draws can and do work, but most individuals simply do not have the patience to get it right.
Most of those who tried and failed with a soft back drawn blade, had blades that broke. Upon examination of the broken cross sections, it was easy to see that the blade had a tempered "skin" of about .010-.020, but the core of the blade was still "as tempered" hardness. After many rounds of questioning, it became obvious to me that the individuals were relying on the oxide color change as an indicator that the tempering was where they wanted it...but they were only achieving that .010-.020 tempered skin. The steel must be at a given temp, for a given amount of time to "temper", and most folks just don't achieve the necessary duration at temp with a torch.

I personally think its way easier, and much more predictable to adjust the depth of an edge quenched blade, and then temper for AT LEAST 2 HOURS.

I know there are others who either do not like, or disagree with edge quenching, but thats a matter of opinion. I like it, I use it, and think that it's far easier and far more predictable than fully hardening a blade and then trying to "soft back draw" the spine(s).

 

[Alarion]

I'm with Ed.
I've done spine tempering with a torch, but you got to keep the edge and point in water, and use a gentle heat for a long time.
You have to heat it up real slow and even. It may take over 20-30 minutes to get to the color you want, and then you have to keep it that way for at least as much time to be sure it's ok.
I do color tempering over the forge but, again, it may take a full 40 minutes to get a deep straw color. Remember that steel is an awful thermal conductor.
In the end, I much prefer to edge quench or heat the steel so that only the edge and point are at austenizing temp, then quench vertically the whole blade.
With the right steel, and proper heat, I managed to get a good temperline this way, and while a file would skate on the edge, it still could bite nice and well in the spine.

 

[Sam Salvati]

If you are trying to pass the abs test then sure Ed's way is good. but if you are looking for a knife that performs better in the real world where a crowbar is used for prying and your knife is used for cutting then a spine of complete martensite that has been tempered more than the edge is tougher than a blade where the spine is pearlite (like on an edge quenched blade).

To actually answer Justin's question, the mapp torch should do the job? I have tempered hammer heads with a plumber's torch. are you looking to get the colors to run? Is the steel polished when you put it edge down in the water and torch it?

Ed, I find it disappointing that on the one hand you hold a good example of time having a factor in tempering, then on the other hold edge quenching as proper practice, but then again it is just a matter of opinion.

 

[Justin King]

Maybe I'm just not patient enough, the water just seems to suck the heat out too fast. I appreciate the opinions. I feel that a through-hardened blade is better suited to most users, and to me drawing the spine is just a way to get a little more toughness out of a blade that should be tough enough already.

 

[JasonByers]

I believe in the real world most who buy and use knives does not know the difference between a cutting tool and a prybar. I also think most folks who break knives in this manner will probably blame the maker. Depending on the steel I always edge or clay quench. Also if you still want to draw the spine you can etch your blade after for about a minute and that will show you if your tempering is in the correct position.

 

[VaughnT]

Ed, are you saying that you put the knife in a pan of water and slide the whole thing into the oven for two hours to soften the spine a bit?

 

[Sam Salvati]

Ed, are you saying that you put the knife in a pan of water and slide the whole thing into the oven for two hours to soften the spine a bit?

water boils at 212F, useless for tempering really.

 

[Justin King]

Ed is talking about quenching just the edge during the hardening cycle, and then using a normal oven temper. This is differential hardening, which is different from differential tempering, the latter is what I was asking about.

 

[koyote]

I've been fully hardening my blades for several years now (still shock myself with that!) and I use MAPP for the spine draw.

Now, a lot of this depends on 3 factors:

steel - 5160, for example, can get brittle if this is done wrong. I'm a lot more comfortable with my (completely wrong and totally not advised, but it WORKS) "spine at the edge of the tank) quench on thick - 3/16 and up- 5160 blades. 15N20, 8670M, and the 1080,1084, and Aldo 1084FG I use all respond fine to the MAPP torch.

Thickness- I do a lot of thin blades, and it's not that bad on the 3/32 through 3/16 spines. I tend to work an inch and a half or so at a time, moving from tang to tip. On the "true 3/16" or 1/4 inch blades it's rough. But doable.

Environment- wind, angle, and lighting are all big deals here. I work more or less outside and even a light breeze will suck your time up.


Performance? I've had *one* broken blade out of some hundreds of knives, and that was a 1/16 blade being batoned through oak the hit a knot. It was an intentional abusive test, and it's not really a surprise. It's common and accepted that people will baton the 3/32 and 1/8 15N20 blades through anything made of wood, in any temperature. Granted, I'm starting with possibly the best material for a thin blades being pounded on with a mallet, but I think the MAPP spine draw works.

Oh- I had a B***H of a time keeping blades perfectly trued doing edge quenches on the thin stock.

I have no issues with edge quenching, and I do some 5160 that way as it is. I'll be doign more once we move and I get sucked into the borg.

 

[Sam Salvati]

I've been fully hardening my blades for several years now (still shock myself with that!) and I use MAPP for the spine draw.

Now, a lot of this depends on 3 factors:

steel - 5160, for example, can get brittle if this is done wrong. I'm a lot more comfortable with my (completely wrong and totally not advised, but it WORKS) "spine at the edge of the tank) quench on thick - 3/16 and up- 5160 blades. 15N20, 8670M, and the 1080,1084, and Aldo 1084FG I use all respond fine to the MAPP torch.

Thickness- I do a lot of thin blades, and it's not that bad on the 3/32 through 3/16 spines. I tend to work an inch and a half or so at a time, moving from tang to tip. On the "true 3/16" or 1/4 inch blades it's rough. But doable.

Environment- wind, angle, and lighting are all big deals here. I work more or less outside and even a light breeze will suck your time up.


Performance? I've had *one* broken blade out of some hundreds of knives, and that was a 1/16 blade being batoned through oak the hit a knot. It was an intentional abusive test, and it's not really a surprise. It's common and accepted that people will baton the 3/32 and 1/8 15N20 blades through anything made of wood, in any temperature. Granted, I'm starting with possibly the best material for a thin blades being pounded on with a mallet, but I think the MAPP spine draw works.

Oh- I had a B***H of a time keeping blades perfectly trued doing edge quenches on the thin stock.

I have no issues with edge quenching, and I do some 5160 that way as it is. I'll be doign more once we move and I get sucked into the borg.

all this leads me to beleive that the torching on very thin stock might have actually caused some minor hardening, 5160 as we all know is fairly deep hardening.

 

[Justin King]

I'm a lot more comfortable with my (completely wrong and totally not advised, but it WORKS) "spine at the edge of the tank) quench on thick - 3/16 and up- 5160 blades.

This caught my interest-could you give a little more detail on this technique?

 

[koyote]

all this leads me to beleive that the torching on very thin stock might have actually caused some minor hardening, 5160 as we all know is fairly deep hardening.

5160 is fairly deep hardening, but it's also a bit too prone to air hardening (in my experience) which is why I'm not using a MAPP torch for a spine draw on it.

Regarding the torch on thinner stock. I'm not sure what you are suspecting. I do a full quench and oven tempering process, then use the torch to draw the spine. Which part fo the steel am I rehardening? And do you think I'm hardening it somehow past the oven tempered hardness?

I should also clarify that I have the edge in water while doing this, as well.

 

[koyote]

This caught my interest-could you give a little more detail on this technique?

Okay- disclaimer. I'm a mechanic, not a scientist. What I'm doing is probably completely wrong. I don't have testing gear that's granular enough to give any answer as to microscopic internal structures.

Kevin always says- test this yourself- And I shake my head and ask myself when I'll have an electron microscope in my living room

Well, here, you need a vise, and stumps, and some scrap wood. So test it yourself!

My longer quench tank is made of a piece of 6 inch steel pipe. welded a bottom onto it (it's about 40 inches) and sat it in a metal 5 gallon bucket. (to catch overflow). I use canola oil for quenching in that tank. (disclaimer)

I preheat the oil to about 130-140. The blade -as stated, mostly with 5160 I'm doing choppers and the blades are in the 3/16 to 1/4 inch thick category- the blade goes in for normalizing and to make sure everything "wants" to sit straight.

With the quenching heat, quenching is done tip first, full immersion, into the tank, and agitated "swooshed" around with the spine being against the wall of the tank.

Then I let it cool to bare hands temp and go scrub it and pop it in the oven. I generally do a single, 2.5 hour temper, at 375-425 depending on the blade use.

Oh- geometry. I'm doing full convex blades with this. I suspect this matters, but a flat or saber ground blade should work out well.

I've been told that I'm hitting 57-58 on the edge and 52-54 at the spine. But I don't have a lab.

I have no desire to do a 90 degree bend in a blade at this point, but I can get 25 degrees each side without breakage or "setting" the blade.

My other tests involve throwing the knives at the tomahawk targets, and chopping random wood- 4x4 or "weed trees". No chipping or breaking during throwing, and no rolling or chipping on the regular valley oak or fir woods.



Disclaimer- I am NOT a master smith. I am not even a journeyman smith. I am not a trained metallurgist. My quench tank is too tall and too narrow and canola oil is too slow. Experiment with what I'm doing, but do not consder me to be an instructional source on heat treating 5160. A: I'm not that skilled (I've 4 years in this, I'm really not that skilled) and B: I'm pretty sure I'm doing it 'wrong'. (But there appear to be at least 9 wildly different ways of doing it 'right')

 

[Alarion]

I believe all this stressing the temper leaves out one important factor.

A couple of years ago I tried my hand at a yari point. Something I wasn't obviously up to.
I made a sad mess of it, the point being ground too thin near the ricasso, and almost twice as thick in the second half of the blade, towards the point.
Disgusted with the result, I was about to toss it away, when a post from Kevin came to mind. I fired up my small gas forge and heated the point.
After normalizing three times, I started the hardening process.
I did INTENTIONALLY leave the point so that while the thinner half heated gently to non-magnetic, the forward, thicker portion heated up much more, far above that temp.
I left it there for some time, quenched in oil, and did a color temper to brownish-purple.
Then I put the TIP of the blade in the vise, put a tube on the tang, and started bending it one side and then the other.
As many can guess, the blade SNAPPED with brittle fracture NOT at the point of least section, near the ricasso (where the bad grinding also left some stress riser corner) but at the THICKEST section.
I put the blade in the vice again, and worked it again back and forth to break the thinner section.
I had to sweat it! The blade was hard to bend, tough and wouldn't break.
I do not posses an electron microscope, but have good sight at close range.
I examined both breaks under strong light.
The thicker, overheated section had a coarse grainy look, looked like a broken shard of concrete.

The thinner section gave with plastic deformation (like when plastiline breaks if you pull it too much). The metal had flowed somewhat in the last bend.
The surface of the broken blade was fine gray, silky, like a broken shard of very fine porcelain, but without the jagged edges.
Before somebody thinks it was pearlite and not hardened, a file skated right over it, and I kept that piece of metal, sharpened it an used it as a scribe. I still have it somewhere in the attic.

So: grain growth is the first and foremost enemy. Even with a color temper, done somewhat in a hurry, the properly treated part of the blade was surprisingly tough and hard.
Where the grain growth was allowed to wreak havoc on the steel, it was a real mess and a very inferior quality piece of metal.

I've witnessed several broken swords from a friend who has a reenactment group and does medieval swordsmanship.
Almost invariably, at the break there was the same coarse grain, shiny "crumbs".
So, I believe that it's true that temper makes or destroys a good blade... provided that a good blade is there in the first instance, and not a piece of hardened "steel crumbs" loosely held together by whatever is that holds together so abused a piece of metal.

My first goal is avoid grain growth. Then I go for a gentle, long temper to the color most suited to purpose and steel.
I really like differential heat treating, I feel it's really THE difference between a custom made, lovingly made blade and mass production.
But I think that, in a normal knife, subject to normal, skilled use, it adds nothing to the blade itself, if not just super-extra margin of safety on the blade's toughness.
Or, to put it another way: it's not strictly necessary, but it IS better, and I know it is there.
The difference between a custom made artifact and mass production is exactly that. Mass production strives for economy.
Material engineering seeks the material that satisfies the requisites with the least expense. Craftsmaship seeks the BEST material on an absolute level, with the BEST treatment you can give it.

 

[Kevin R. Cashen]

... but if you are looking for a knife that performs better in the real world where a crowbar is used for prying and your knife is used for cutting ...

Hehe, with a computer in my lab just beside the forge room, it is always fun to see if I can slip in a post between welding heats

Sam there is one problem with your argument…

Crowbars are not supposed to bend!

I have been at this for 30 years and outside of the ABS test I haven’t been able to find any reasonable advantage to a knife that bends, or a crow bar that would bend, nor a spring that would bend. All of these objects balance toughness with strength, not ductility.

 

[DrThunder88]

5160 is fairly deep hardening, but it's also a bit too prone to air hardening (in my experience) which is why I'm not using a MAPP torch for a spine draw on it.

Regarding the torch on thinner stock. I'm not sure what you are suspecting. I do a full quench and oven tempering process, then use the torch to draw the spine. Which part fo the steel am I rehardening? And do you think I'm hardening it somehow past the oven tempered hardness?

I should also clarify that I have the edge in water while doing this, as well.

Is it rehardening or just becoming embrittled? I think brittleness increases above about 500 degrees F even though hardness decreases.