totally confused, whats your opinion ????

B . Buxton

B . Buxton

Joined
Jul 8, 2001
Messages
3,623
I've been cruising the forum and talking to other knifemakers that forge and heat treat their own blades, and I have been told different answers about this process. I have been doing this for quite a while now, but I am always willing to change my techiques if it will result in a better finished product. This is what I do and have had very good results with my testing. I would like to know if this is on the same line as other makers or if I can tweek it in a different way or what. First after forging, I bring my blades to non-magnetic and let air cool to the touch 3 times, on a 4th heat to non-magnetic I place the blade in a bucket of wood ashes, that has been pre-warmed with 2- 2" x 8" x 1/4" pieces of plate steel heated to orange hot. After several hours of cooling, I remove profile and grind the blade, next I heat my quenchant to the temperture I want to within 5 degrees, bring the blade to non-magnetic and quench edge first, after the blade quits bubbling I submerge the entire blade in the quenchant until cool to the touch, I do this 3 times leaving 24hrs between quenches. Directly after the 3rd blade quench and cool down I do my tempering cycles of 3 - 2hour session. I pay very close attention to my quench temp and blade temp(with a magnet) before quenching.
I know that ways of improvement are discovered everyday, and that we all have are own routine, I'm just a little confused after talking to a couple other makers and there fore self doubt sets in.
:confused:

Thanks
Bill
 
Bill,
This is what little I know, which ain't much.

Forge
Normalize
Anneal
Rough grind
Heat cycle (3 times)
Heat treat
Temper (draw)
grind and sharpen

I am far from being an expert and this is only the basics. Maybe someone who knows the "who's what" and "where fors" can chime in.

I have been told the exact same method that you use before but the heat cycling after annealing just made more sense to me. The reason for the heat cycle AFTER annealing is to have a tighter or finer grain. Annealing promotes grain growth.

C Wilkins
 
Bill, you didn't mention what steel you are talking about. If you are using 5160 or 52100 your heat treating method sounds real good. I've had pretty good luck with this method (triple quenching) myself on the aforementioned steels. The only thing I would do would be to add a round of normalizing heats after grinding the blade and before the first quench.

If you are using steels like the 10xx series or O1, W1 etc. then the multiple quenches are probably a waste of time. (That however is based solely on what others have said and that is that these steels appear to derive no added benefit from multiple quenches.) I've only used the triple method on 5160 and 52100 because of the research of the effects of this method done by Ed Fowler and Ed Caffrey. The conclusion from their test and using knives is that the working characteristics of these two steels are greatly enhanced by this process.
 
Bill, I do everything the same way you do, except I only do the triple quench for 5160 and 52100. From my readings and the recommendations of other makers, not my own experiments, it appears that other steels such as 1084, 1095 and O1 do not benefit from the multiple quench as do the two other because of the chromium content. I always triple temper. I am not convinced that O1 will not benefit from the triple quench since it also has chromium, although less than 5160 or 52100. However, I have not had time to experiment with this and do some extensive testing. It is on my list of "things to do next". I am pretty sure that Ed Caffrey does 5160 and 52100 the same way you describe.

Fox
 
I guess you're talking about 52100 steel aren't you Bill? That's pretty much what I do with a few variations. So that brings me to my question. Why the 24 hour wait between the triple quenches? Everything I've learned about heat treating tells me to temper the blade ASAP after the quench. Waiting can cause stress cracks during the martensite transformation. Could it be that that just works with 52100 because of some property in the steel? I don't know but it seems to me that you've been awfully lucky so far.

Martensite expands while it's forming. Sort of like ice in ice cube trays. And it can definately cause stress cracks throughout the hardened area in my experience. Of course, I deal mainly with 1084 and 5160 on a routine basis. 52100 is a once in a while thing for me. But I do like that steel.

Here's a story that helps me remember to always immediately temper my blades after quenching. When I was going to the ABS school I was lucky enough to be able to attend a club event at Uncle Al's Riverside Machine in DeQueen, Arkansas. No names because the feller in question might come pound on my head :)

Anyway, the demonstration was the quenching of an oriental style sword blade with clay on the spine for a nice hammon. The blade was straight but the demonstrator assured us that it would curve up in the traditional shape after quenching. Well, it did and that's a fact! He quenched it in preheated water after heating it up in one of Uncle Al's working gas forges. I believe the blade was 1060, but don't hold me to that. Anyway, he did the quench and we all followed him up front to the better lighting around the office area. Ping! Ping! The blade started to crack. And it continued to crack in a line following the clay hammon until it was a sorry mess! I felt so sorry for him and his ruined sword because I knew how much work he'd put into it. It didn't seem to bother him much becuase this was the third time he'd quenched this particular blade. But I knew it was killing him. It sure would me because I've been there a time or two with regular sized blades. That's why I asked about the waiting period between quenches. It just doesn't follow for me. But I like to learn so lay it on me!
 
C. Wilkins, the way I was told and have been performing is that the annealing process is the slow cooling process after reaching non-magnetic, and thru this slow cooling the grain is refined to a smaller state, that normalizing (non-magnetic and air cooling) is done more for a stress reliever of the steel after the forging process.
Max, the reason for leaving the blade for 24 hours is to slow cool in the quench to room temp, I know this doesn't take a full 24 but usually it will take several hours so I always give it 24 or close to it, the slow cooling in the quench appears to act like an annealing cycle to the whole blade that inturn helps refine grain size. I've never had a blade crack working it this way (Knock on wood), but I had some different opinions and that has gotten me thinking (that is always dangerous) and self doubting, which is a knifemakers worst enemy.;) By the way the steel is 5160,52100.

Thanks for you alls advise, I'm getting my confidence back.

Bill
 
Bill, I triple quench but do it as soon as it is cool. I do temper at least two times and most of the time three. After a temper cycle I let it cool to room tempature. That takes a half hour at most. It goes in the freezer for 24 hours. I do this with each cycle. A very fine knifemaker told me his knives held a better edge in winter and he started freezing. It does make a difference, not only in the edge holding but in the cutting edge the knife takes in the first place. I finished a little kneck knife todday. You could not find a straight razor with a keener edge.

This process also seems to make the steel flex better. I used to temper 5160 at 325. I now use 300 and it does better on the brass rod test. The steel is harder and you can tell during sharpening but it os not as brittle.
 
I havn't got as much experiance as a lot of smiths, but I have been testing and trying a lot of differnt methouds.

This is what works for me, I forge in the 15-1600 deg. range, after forgeing, normalize 3 times in still air, aneal 3 times in wood ashes 24 hours appart(don't know if it's necisary or not, I just aneal at end of day), rough grind, triple quench at least 24 hours apart(leting the blade set in the oil), set for a couple of days in the frezzer, then triple temper at 375, check with a brass rod, if it dosent' chip out or isn't to soft I then finish it.

I have found the 24 hour seperation to be critical for me. I once cut it short on a blade( I figured about 16 hours, thought that would be enough) and woundered why it didn't cut that well. Normalized it 3 times and started over, with 24 hours between quenches, and it cut like it was suposed to.

I have only had one blade crack, and I think that was due to my first time useing a spring fuller, and fiddle farten' around to long and let the metal get to cold to hammer.

I am interested in anything else I can do to increase performance, and am always trying new things.
 
Bill,

It looks to me as though you are doing everything right. The only thing that I do differently is that I give the blade three annealing heats and I wait twenty four hours between the last quench and the first tempering heat. Also it seems to help the cutting to put the blades in the freezer after they reach room temp. With 52100 Ed and I have found it unneccessary to go all the way to nonmagnetic for the annealing process. In testing, 1100 has been plenty hot enough. I quench everthing that I make three times and also temper three times. I recently used this method on some stainless damascuss and found some significant performance gains. The blade that I heat treated in this manner cut about three times better than the reported normal 40-50 cuts on hemp rope. Also I know that steel expands on hardening because of the martensite transformation But I have traced around a 52100 blade before hardening and then again after each quench. what I foun out is directly opposite of what is supposed to happen ie. the blade tip dropped! The first quench caused the most significant change with the second and third quenches causing only about half the change of the prior quench. any way this post is getting quite long so to recap:

three normalizing heats; heat to nonmagnetic let cool to room temp.
three annealing heats; heat to at least 1100 deg. cool as slow as possible. wait twenty four hours between cycles.
profile and rough grind. finnish to 240 grit minimum
quench three times. wait twenty four hours between cycles put in freezer between cycles.
temper three times. wait twenty four hours between cycles.
finnish grind and test. brass rod test then cutting tests.
if the brass rod test shows that the blade is too hard raise tempering heat twenty five degrees at a time until the blade will pass the brass rod test.
 
I am proud of the thoughts expressed in this thread. One time we would have been considered heritics.
My thoughts.
Blades that crack when hardened were possibly:
Hardened in oil that was too fast for the steel.
Over heatred while forging or when quenched.
Errors in the forging process create faults that seperate during the guench.
Some forging profiles are especially prone to failure during the quench.
Steel that had some quality problems.

The 24 hour wait and time in the freezer combine to develop toughness. Most of what is going to happen to a blade hapapens quickly, the remaining changes are slow to mature. Infinite changes happen throughout the life of a blade. The 24 hour pause allows enough time for the vast majority of what we want to happpen, to happen.

I explored the above for over 20 years and each thought or theory if you wish to call it that is based upon many test blades that were reduced to scrap.

I sincerely appreciate your interest in furthering the performance potential of the forged blade. One thought, look at all blades of history and ask such questions as Why and What for? Blade smiths of the past have explored our frontiers very thoroughly.
 
Ed, you'll get a kick out of this. I still have people say "You believe all that hogwash!" when I mention the triple normalize/triple quench/triple temper method. I, however, am a firm believer!
 
Triple normalize and maybe triple quench make some sense but triple temper goes THz, what's the rationale there?

If you can refine grain size with repeated normalizations how many still have an effect, or is it asymptotic? What little searching I have done points to thermomechanical refinement in the industry, anyone tried that (wootz workers excempted);>)

As an engineer am the last person to be against quality empirical evidence but it would still be nice to know why.

TLM
 
No single event done right makes a huge difference. All together combine for the final product. Three seems to be the magic number.

We have a tremendous opportunity, Rex Walter the man of science who has worked with us of over three years has a thread, on the custom forum titled "Etches and the People who Love them" Visit the thread, ask him questions and he will be able to porvide a lot of information.
 
TLM: You asked why the tripple temper. This is why: I was trying to push steel to the absolute limit if cut. Blades are tested for chip before cut. One time I tested both at the same hardness. The blade cut very well, but chipped slightly. Rather than proceed as usual and temper at a higher temperature, I tempered a second time at the same temperature. The edge did not chip as badly as the first flex test. I decided to temper again at the same temperature. It quit chipping and cut about the same as it did the first cutting test. When I temper at a higher temperature I lose too much cutting performance. Therefore when pushing steel to the limit, and I find an edge that chips after three tempering heats, I raise the temperature only 5 or 10 degrees depending upon the nature of the chips each time and temper three times. To tempering temperature in one hour, hold for two and let it cool down slowly in the oven and then in the home freezer for at least 12 hours. Twentyfour hours between tempering heats. Then test again for chip.

This method is not necessairy unless you are trying to push the steel to the absolute maximum level of performance. There may be quicker, easier or more economic methods, This works for me in concert with all the rest of the stuff I do to the steel. Thanks for a very good question!
 
Thanks Ed. This is a very good example what I ment, I can imagine ways for normalization and quench to improve properties (refined grain size) but the temper is more difficult, it should be a diffusion controlled process, so you should be able to play with time and temp. to do it at one go but then again one might be able to have more control by keeping the temp. down and doing it three times.

Nothing I have ever seen written anywhere says that this should work, for some reason I have apparently not read everything ;>)

One can easily see there still hope for the individual to improve things but the way (I guess) is mostly trial and especially (sigh) error. No academic researcher in his right mind would try three tempers or quenchs. (A slight exaggeration as triple temper is used on tool steels to get rid of retained austenite.)

TLM
 
Bill, if you still feel confused, at least you can say that you are confused at a higher level now.

Usually not so seriously

TLM
 
TLM: When I first started makng knives seriously I picked up some metalurgical textbooks. I made up some flash cards and studied hard. Soon I relaized that most of what I sought was not directed to knives, but crankshafts, wheels, and other stuff. I tossed the books into the slough in front of my house and started manipulating variables on steel seeking to find what contributed to steel as knife.

The most important aspect of my venture centered around testing knife performance. This is the most single important aspect of the growth of the maker, how and why he tests his knives. The more reliable and valid his tests, the better his knives will become.

Most if not all of my achievenents came about by accident. A simple question 'Why not try this?' No exact design, just seeking answers when opportunities presented themselves. For every success, there were many failures.

Today I have many more questions than I have answers.

For example, many say multiple quench doesn't work on some steels. I plan on finding out if this is so and if so why. Several hypothesies roll around in my mind, any one of which could keep us busy for some time.

I believe that the absolute antecedent condition to high performance is the degree of thermal mechanical reduction in mass prior to final thermal treatment. The test blade discussed in the April Issue of Blade was formed accompanied by at least a 98 point reduction in mass. Rex has a line on some 15 1/2 round bars we plan on forging into knives, using low temp forging all the way. This will be an informative set of test knives.

I believe that the academic study of some events is limited by the knowledge that places the student in a position to do the research. He already know what won't work and proves it.
 
Ed, textbooks are by definition just that, the cutting edge of research (pun intended) can be found at various papers scattered around, and only one very small piece at the time. Textbooks are necessary to understand the "real" research. You can actually find very interesting bits of information ,it just takes a lot of time and patience.

This site http://www.msm.cam.ac.uk/phase-trans/bainite.html might interest you, this guy is using some deceptively simple steels to get very impressive results, the design methods he uses are anything but simple. You might look for more of his articles.

Thanks for sharing the practical information and techniques with us, I just can't help speculating on the why!

TLM
 
Thanks Ed. That bears out my observation that using the multiple temper/freeze cycles, I can temper at a lower temp and have a better cutting blade.
 
You must log in or register to reply here.
Back
Top