Heat treat testing

J

Jason Fry

Joined
Jun 5, 2008
Messages
3,224
In reading the steel snob thread, several makers talk about their ability to fine tune a heat treat process over several attempts. One maker claimed that in four or five tries, he could develop a "good" HT for a batch of mystery steel. Is there an accepted process for developing a HT? What tests are used during this process to determine if the HT is "good" or still needs work?
 
The proper process is to set up samples to test many variables .Each sample is then tested for hardness HRc], toughness [Charpy test], and the microstructure examined.
Some may not have those tests avilable then it involves some guessing !
 
It is also entirely subjective to what the maker himself wants, I insist on performance in several of the rigorous tests mete mentioned while others are happy with the appropriate file skate, edge flex, or cutting performance of their choosing. One of the reasons I moved to the tests mete described is to remove this relativism, when I speak of maximizing the steels potential it is often entirely independent of individual ideas of knife performance, since I cannot tell anybody else what their standards in that area should be. What I can do is examine totally objective numbers that measure hardness, toughness, or establish exactly what was created inside the steel in its treatment. Some makers think that the results of pearlite are great in a blade, others see it as very flawed, I prefer to show exactly how much pearlite versus martensite there is, where it is, and let each maker take it from there, as long as they understand what it means.
 
If your just starting out a good process for ht for a mystery steel would be to quench your blade in a thin heated oil and temper.Do some cutting tests related to the intended use of the knife.If the edge chips ,temper longer.If it wont hold an edge like you want you will have to get a faster quenching medium like parks quenching oils.The real quenching oils can make mystery steels much more useable because using a steel like 1095,you probably are not going to get much out of it in terms of hardness without it.
 
I did some reading RE: Charpy test. Since it's qualitative, comparing one piece of steel against another, if you used the same pendulum, same sample size, and same measuring standards, it seems like a person could actually rig up a crude tester that would provide qualitative comparisons. I assume there would be several variables within the heat treating process, including (but not limited to): hold temperature, hold time, temper temperature and time, and quench medium.

That said, so far I've been using a process similar to mstephen. I don't have great temperature control, so my results have been somewhat spotty. Even so, I do the best I can with what I've got. I start with the usual "nonmagnetic, quench in oil" using heated vet grade mineral oil. I start with a temper at 400, then test the edge. If it chips, I bump up to 425 for the next cycle and test again. So far I've found out several things about the two major kinds of mystery steel I have been using. One, the trap springs I have used do get hard, but 400 is too high to temper, as they seem to lose their edge holding ability if tempered much higher than that. Two, nicholson files will chip on the edge every time if you temper at 400, but usually not at 425. I hate for the ultimate answer to be "spend hundreds or thousands of dollars on equipment," but at the same time, all of this education (thanks Kevin et al, sincerely) is making me less and less satisfied with the tools I have at my disposal. I suppose if at the end of the day I can make a knife that serves my purposes, and can repeat the performance again the next time, I've made some kind of progress, considering that when I started, there were no guarantees that the knife would perform as I wanted, or that I could do it again the same way.

I do know that I certainly have a scientific mind on a backwoods budget when it comes to this knifemaking stuff.
 
To quote Kevin from another thread:
"Newbies only seeking validation wear me ragged"

I hope that by stating what I'm already doing that I don't appear to be seeking validation. I only posted my current process so that the more knowledgeable could point out where I can improve.

Not to knock mstephen, whose advice I am coincidentally following, and not to disrespect mete and Kevin, whose reputation is well deserved.

Now, anyone else care to answer my question?
 
I know you want a response from others but you should ditch the magnet for telling you when to quench.
 
jkf96a said:
In reading the steel snob thread, several makers talk about their ability to fine tune a heat treat process over several attempts. One maker claimed that in four or five tries, he could develop a "good" HT for a batch of mystery steel. Is there an accepted process for developing a HT? What tests are used during this process to determine if the HT is "good" or still needs work?
Click to expand...

I'm surprised nobody has mentioned the real world physical performance tests yet. Even the ABS uses physical tests to test a makers ability to make an acceptable blade and the cut tests are used before the destructive tests.

I believe I've read that one of the tests is to cut a 2x4 twice, then still be able to shave fine hair on inner arm or push cut newspaper or something... maybe somebody that's been there can share the actual spec.

There are other tests such as how many times you can cut a 1" manilla rope before needing to be sharpened.

These tests will not only test your blade and HT but your ability to put an 'adequate' edge on your blade. Think convex.
 
jkf96a said:
I did some reading RE: Charpy test. Since it's qualitative, comparing one piece of steel against another, if you used the same pendulum, same sample size, and same measuring standards, it seems like a person could actually rig up a crude tester that would provide qualitative comparisons. I assume there would be several variables within the heat treating process, including (but not limited to): hold temperature, hold time, temper temperature and time, and quench medium.

That said, so far I've been using a process similar to mstephen. I don't have great temperature control, so my results have been somewhat spotty. Even so, I do the best I can with what I've got. I start with the usual "nonmagnetic, quench in oil" using heated vet grade mineral oil. I start with a temper at 400, then test the edge. If it chips, I bump up to 425 for the next cycle and test again. So far I've found out several things about the two major kinds of mystery steel I have been using. One, the trap springs I have used do get hard, but 400 is too high to temper, as they seem to lose their edge holding ability if tempered much higher than that. Two, nicholson files will chip on the edge every time if you temper at 400, but usually not at 425. I hate for the ultimate answer to be "spend hundreds or thousands of dollars on equipment," but at the same time, all of this education (thanks Kevin et al, sincerely) is making me less and less satisfied with the tools I have at my disposal. I suppose if at the end of the day I can make a knife that serves my purposes, and can repeat the performance again the next time, I've made some kind of progress, considering that when I started, there were no guarantees that the knife would perform as I wanted, or that I could do it again the same way.

I do know that I certainly have a scientific mind on a backwoods budget when it comes to this knifemaking stuff.
Click to expand...

You could have your mystery steel tested.

There is nothing wrong with non-industrialized HT-ing. All the equipment is for consistency. Without the equipment, consistency has to come from method alone. The absolute best way to not have equipment, yet have best-consistency-opportunity, is to use known steel... it provides a base to work method on.

If you can do relative hardness testing (Wayne Goddard did an article in Blade Mag. on it recently), mystery steel and consistency get easier.

Edge chipping or flex/return or bend does not tell you how hard steel is. I have a knife with a very thin blade and very pointed tip. It is 63-64HRC on a calibrated Rockwell tester. It has a bent tip. It needs to be taken apart and reheat-treated harder because it should flex/return... right?

A person can "back woods" but it really helps to know metallurgy... it's a consistency that can be counted on like known-steel... results that don't match consistencies is improper method. Now that's science... =]

Mike

Mike
 
jkf96a said:
... I hate for the ultimate answer to be "spend hundreds or thousands of dollars on equipment," but at the same time, all of this education (thanks Kevin et al, sincerely) is making me less and less satisfied with the tools I have at my disposal...
Click to expand...

Allow me to break the good news that I do not believe that is the ultimate answer. I have spent quite a few years assembling what amounts to a metallurgical lab, and it is well above anything a bladesmith requires to make good knives. Since I have it I do apply it to my knifemaking but my motivation in collecting my analysis equipment was for more advanced research in order for me to back up my teaching and writing with solid information while taking on so much of the nonsense that is backed with none. Some folks have the most recent knife magazine and their "experts" opinion as to what is going on in the steel, I have actual images of the inside of the steel. I don't need to ask other bladesmiths, I get my facts straight from the source- the steel.

For making your own knives and trouble shooting them it is really just a matter of knowing what you are testing for and what your tests are really telling you, for this your absolute best tool in knowledge about the properties you are looking for. A file will indeed tell you a good part of the story on heat treatment but if you can get access to a hardness tester you can get both sides of the hardness equation, scratch and penetrative hardness. So you will know your knife has both edge strength and abrasion resistance. I strongly advise caution in putting too much in any test that involves flexing, since so many makers (even well known ones) seem to have a shockingly limited understanding of what that test can actually tell you.

Impacting the edge is good in determining toughness, and a section of brass rod is all you need to check it out. A good edge should be able to handle some moderate blows to the rod when sharpened. Next you move onto actually just using the knife for what it was intended for. I am not big on overall knife tests that involve things the knife will never really do in actual use. Use it and see how it works. Use it for a long time and see how the handle feels. I like chopping barn beams and clearing fence rows to test both the knife and the handle design.

So there you are, a file, a brass rod, various wood items to chop and you can get a decent idea of where you are going, and if you want to really narrow it down a hardness tester is a worthy investment, but nothing that will require a mortgage.

Now with more sophisticated equipment your process will become more predictable and consistent and your testing can be less rigorous in order to see where you are at at any given time, but that is where not knowing the exact chemistry of the steel will also be a limiting factor. There would be no sense in an oven or salt bath that can hold an exact temperature if you have no idea what the temperature should be. One common theme among users of scrap or reused steel is a strict adherence to continual testing with every blade, this is necessitated but the fact that each result could be different. After assembling the shop I have aspired to over the years, I only test a small sample of the steel from the same stock all the blades in a batch are made from that has accompanied them through every step. It is a very standardized method that does not tolerate wild variables or deviation, so you can see why my world runs best with fresh steel with specs and chemistry.
 
You must log in or register to reply here.
Back
Top