View Full Version : Flash tempering

Sean Cochran
11-29-2011, 08:26 AM
What are the drawbacks? (sorry bad pun...but I couldnt resist :o)

I have always tempered in an oven, but I know some guys who flash temper, or temper with heat and watch the color until it goes to straw.
Is tempering a matter of just heat or is it heat and time?


11-29-2011, 09:56 AM
Not sure how some folks do it but, traditionally the japanese put thier swords directly into the coals until blade temperature reaches 300 degrees f
they dont want to ruin hamon with too much heat . after all you dont want to heat it up a lot as the blade ( spine is only about 40 RC ) if done too long it will make spine too soft . This the way I was taught . I am sure there are other methods .

Doug Lester
11-29-2011, 11:03 AM
From what others have said the big problem with flash or draw tempering is that you may not adequately temper the steel all the way through. It can also be tricky to judge the colors. The temper could end up uneven, but that could actually be turned into an advantage if done by design. The big advantage is that it's a whole lot cheaper than a 36" deep tempering oven.


11-29-2011, 09:02 PM
Terminology question if I may. I think of draw temper as painting on a temper with a torch - probably to soften the spine while leaving the edge protected. I also have heard draw as a synonym for tempering. When I use the term "flash temper", I'm talking about a quick - relatively low temperature temper that may be used take the stress off a blade (maybe in the kitchen oven) if the kiln is going to take hours to cool down enough for proper temper. Similarily, a flash temper may be used to take stress off the complex hardened parts before going to cryo.

I get the impression that "flash" is being used differently here. Can someone explain its context here?


Sean Cochran
11-30-2011, 06:48 AM
Terminology question if I may. I think of draw temper as painting on a temper with a torch - probably to soften the spine while leaving the edge protected. I also have heard draw as a synonym for tempering. When I use the term "flash temper", I'm talking about a quick - relatively low temperature temper that may be used take the stress off a blade (maybe in the kitchen oven) if the kiln is going to take hours to cool down enough for proper temper. Similarily, a flash temper may be used to take stress off the complex hardened parts before going to cryo.

I get the impression that "flash" is being used differently here. Can someone explain its context here?


What I mean by "flash temper" is tempering relatively quickly with heat. I have seen it done by holding a knife in front of a propane forge (a la Wayne Goddard) or by laying the knife on a metal plate and heating the under side of the plate so that the heat is transfered to the knife. Using a torch is most likely another way to flash temper. When using this method the color of the steel is used as a guide to know when the temper is complete, that is, when the steel reaches a straw color it is done.
This method is in contrast to a "normal" temper, where the blade is placed in an oven, kiln etc. at a certain temp for a given amount of time.
I should mention this would apply to "non-stainless" steels such as 1084, 1095, 5160 etc.

I realize that tempering is intended to relieve stress and soften the blade somewhat thereby increasing toughness.
The question I have then, given that flash tempering appears to work, is tempering a matter of heat + time or heat only?
Or is flash tempering generally not a good method? My thoughts would be that it may not get a full temper to the entire edge which leaves portions of the edge prone to chipping.


11-30-2011, 08:32 AM
I had a guy show me how to flash temper blades by using a torch to heat the spine at the ricasso, until you could JUST see the steel getting brown/dull red. Granted this had to be done in a dark shop, so you wouldn't over heat.

Certainly not the most scientific method, but I have used it a few times on some small neckers and what-nots. Always with pretty decent results. I am not sure about trying to do something bigger than a 2" bladed necker though.

11-30-2011, 08:35 AM
It works well on japanese blades that have been clayed . They have used this method for about a thousand years. Not sure how it would work on monosteel blades with no Hamon ? Its main focus is to maintain the hamon , while not softening the spine to much . Right out of quench the spine is around 40 rc or so . Bubba

Justin King
11-30-2011, 08:49 AM
Flash tempering probably combines better with a partial bainite or pearlite microstructure. My guess is that the closer you get to 100% martensite structure, the more important the full time/temperature effect is for proper tempering. Remember that not all austenizing/quenching procedures give the same microstructures, or in the same proportions, and the purpose of the temper is mostly directed at unstabilized martensite. This is straight off the top of my head though, and I could be in left field.

11-30-2011, 09:01 AM
Ummm... what? LOL

Sean Cochran
11-30-2011, 12:56 PM
Bubba, If im following correctly, the reason for that style of temper is to make sure the blade is tempered "just enough" so as to not be brittle and not lose the hamon with excess heat buildup?

My question I guess is do both methods acheive the same result. In other words does the micro structure of the steel respond the same to heat only as it does to heat + time?

You out there Kevin?

11-30-2011, 01:10 PM
My question I guess is do both methods acheive the same result. In other words does the micro structure of the steel respond the same to heat only as it does to heat + time?

To the best of my understanding, no. Time is required, in addition to heat. I do not know the metallurgical reasoning behind it, only that my own testing has confirmed it. I suspect the thinner sections of the steel temper faster than the spine.... more time is required to sufficiently temper the spine, especially in a thicker blade. In my own testing on a fully hardened blade, the edge held while the spine fractured unless the blade was held at tempering temperature for at least 45 minutes... I temper for two hours, at least twice, to be on the safe side, even if I do a soft-back draw.

11-30-2011, 01:16 PM
yes kevin, here is what I know Yaki-modoshi: Blade Tempering

Due to stresses induced during rapid cooling of the red-hot steel, less than 50% of blades survive the yaki-ire process without cracking. Those that do survive must then be tempered (stress relieved). During tempering, the blade is heated again, this time to a much lower temperature (approximately 300 F) and re-quenched. This lower temperature, since it's well below the critical temperature, won't alter the molecular structure of the steel. Instead, it will simply help to relieve any internal stresses which have built up. Excess heat will affect the hamon . the more time blades are subjected to heat the softer they become . Since the spine of a japanese sword is relatively soft around 40 rc, too long exposure will make spine softer making them more likely to bend . I hope I explained this properly . This is how I was taught when I lived in Japan . although the five main schools called the gokaden, all did thing a little differently. There was some dissagreement among the ancient makers and they would actually resort to armed conflinct to prove who was right . Funny how some things never change.......... Bubba

Sean Cochran
11-30-2011, 02:30 PM
Let me just say I hope I havent stirred a nest of bees. :)

What Bubba says would lead one to think that temper is a matter of heat only, being that the blade is heated to a specific temp and then quenched presumebly to keep the heat from climbing. No doubt this works or it wouldnt be practiced.

I think there is some validity to what GHEzell has pointed out as well, that is, thicker sections would require more time to reach a temper because the mass is greater in those sections relative to thinner sections. However if temp is the cause of softening of the steel (or stress relief) then it stands to reason that the time involved is needed only to bring the thicker portions up to a specific temp. If that temp can be reached in a shorter amount of time (i.e. bubba's example) is the end result the same?

Just for clarification, I do temper in an oven to a specific temp called for by the steel, for an hour X2. I do that because thats what I was taught. Im just wondering out loud as to why that is the correct way metallurgically speaking.

11-30-2011, 02:33 PM
One thing I get questioned a lot about is tempering. I was tought to use a torch to temper. After heat treat I clean up the blade to the point I can see the colors. I heat the back of the blade with a rose bud tip until I see straw. Then I keep adding a little heat until I get the peacock purple. I am told this is "spring" heat. Is the correct? I dont know, but it works for me. I have bent some of these blades and the spring back. I have never had one break.

Some ABS minded people would not touch one of my blades because they are set in their mind this is not the correct way to temper. Hold on folks! 75 years ago the old blacksmith would do this in his forge. He did not have any fancy ovens and the stuff he did has held up! I though the ABS was founded to preserve the "old way" of making a blade.

12-01-2011, 06:34 AM
speaking from my own experience , I believe there is more than one way to skin a cat. In days of old , all blades were fire tempered . But you are talking simple carbon steels . With the advent of highly alloyed steels ,heat treating took on a different meaning. like you said , your friend did it without fancy HT ovens. 1060 responds very well to fire tempering as do other simple carbon steels. Contrary to what abs teaches, you can oil quench a whole variety of stainless steels. 440 C works well with oil quench. ovens are more consistent for HT . But with practice almost any steel can be oil quenched . fire tempering highly alloyed steels is another question. never tried it . who knows it may work ?? The eastern cultures fire tempered all their wootz steels as did the japanese with their tamahagane ............. lots of those techniques are long forgotton.
modern smiths dont know a lot of those ancient ht formulas or recipes as I call them ..... Bubba-san

Justin King
12-01-2011, 07:56 AM
What I was trying to say is that a 1095 blade that is taken to critical and quenched immediately in used motor oil (without a soak at temp) will have a microstructure that is partly pearlite and partly martensite. This microstructure might very well achieve a useable level of hardness/toughness with a fairly minimal temper, such as a flash temper as you describe. The same piece of 1095, given a proper soak at temp, and quenched in Parks 50, will have a much higher percentage of martensite, and I think will not be sufficiently drawn back by a flash temper. In the latter case where the austenization and conversion to martensite is more complete, the combined effect of time and temperature during tempering is much more important. It might be worth noting that the second blade in this scenario should come out of the quench at higher hardness than the first.

12-02-2011, 10:52 AM
Heat treating clayed blades is a whole different ballgame than full hardening of blade. I agree if its a full hardened blade. On a differentially hardened blade quenched in water, will be extremely hard while the spine in much softer , allowing the blade to be used at a high RC because the softer spine will protect the hard edge by flexing some . When doing hamon I always clay the complete blade including tang, less the Ha ( edge).
so the need for heat and time is not so critical on Japanese blades. On a fully hardened blade you are correct. Lots of makers dont clay the tang ,claying tang keeps spine and tang flexible. The application of clay was a technique to make a flexible blades, the hamon is a secondary result. A water quench will fully convert the pearlite to martensite in most carbon steels . Hardness depends on quench medium and the steel and how long temper cycles is etc: When I do a full hardening of a blade I do it the normal way . as you suggested. Depending on what blade is used for , harder is not neccessarily better At the same time 2-3 tempers are not always critical... There are a lot of variables ........... Bubba

Doug Lester
12-02-2011, 07:10 PM
Bubba-san, I think that you might have misspoke. Pearlite will not convert to martensite. Austinite will convert to martensite if cooled quickly enough and the speed needed to convert austinite to martensite will depend on the alloy that you're working with.

Tempering depends on carbon migration and the speed of migration depends on temperature. The distance that it will travel will depend on the speed caused by the temperature and the distance that it need to travel. This comes into play most when it comes to welding up steels of differing carbon contents. With tempering the carbon only has to escape the iron matrix. From practicle experience, time is a factor in tempering but temperature is the most critical, I'm just not versed in steel metallurgy enough to say why. Far more experienced smiths and ABS judges have noted that blades with inadequate tempering time have a higher rate of failure when it comes to the bending test.


12-03-2011, 06:46 AM
Two things with steel that are near impossible to separate are TIME and TEMP. For matrix changes to occur, it requires a given temp, for a given time. An example I can offer is when folks try to do a "soft back draw"..... very often they attempt it with a torch, or with a heated bar of steel, and rely on the "colors" to determine when they are done.

Many times over the years I have had folks who were testing for their JS or MS who tried to use the method(s) described above. In a high percentage of those cases, their blades broke when attempted the 90 degree bent portion of the test...after examining the broken pieces of blade, it's very obvious that there was only a tempered "skin" of about .010-.020" on the blade(s)......they got that little bit of the exterior surface tempered, but beyond that they blade was still at full hardness. (which means the "core" was not converted)

I've done a great deal of experimenting with this over the years, trying to "speed up" the process.

I'm not saying it cannot be done, but it takes TIME at the given TEMP, to achieve. You cannot count on the oxide colors that are visible on the blade's surface. In short, after a lot of examination and experimentation, in general, it takes placing the blade's edge in water (of something to keep the edge cool), and AT LEAST 40 mins of heat application to get the steel's matrix to change all the way through a .250" cross section..... that may not be a "scientific" answer, but that is how it works in a practical application, and I make the assertion based on experience.

Everyone must make their own decisions, but again, it's nearly impossible to separate the TIME/TEMP relationship when it comes to steel.

Doug Lester
12-03-2011, 12:22 PM
Ok, I hit the books last night on this. The text spoke of stress relief from tempering but didn't get specific about the stress but some if it would have to be the stress on the atomic bonds between the iron atoms caused by the trapped carbon atoms within the body centered crystal. It states that the majority of the stress relief will occure within the first 2 hours. So it would seem that yes you could get some stress relief in a matter of seconds or minutes but for full stress relief you will need much longer than that. It will also depend on how the the heat is applied. A low temperature applied over a longer period of time, like resting the spine on a large hot bar would allow the heat to be applied longer than with a rosebud tip on a torch. You could also apply the draw tempering for many cycles instead of just a few.

As far as what was done "way back when" you also need to consider the steel that they were dealing with. From what I have been given to understand from people who have made and used tamahagane and similarly made steels, that it is shallow hardening to the extreme and you may not get full hardening with the steel to begin with. You just don't have enough time to completely beat the nose of the curve to avoid any formation of pearlite and bainite. It like the situation with clay coating, you can use it with 1050 but it will probably not be useful for differential hardening with 6150.