Edge Quench

[Dan Hockensmith]

Ed, and anyone using Eds method. What qualities do feel are gained by edge quenching with the rocking, rather than quenching the whole blade, seeing as only the edge has been brought up to temp? Dan    

[_MaxTheKnife_]

Hi Dan. Once I discovered the edge quench, I totally abandoned the full blade quench. For several reasons. But since you didn't mention the type of steel you're working with I can't really answer your question. I watched Ed's 52100 video and enjoyed it thoroughly. But Ed's way of edge quenching is not mine. And that's ok. Ed achieves wonderful results that are run through the gamut of testing Ed does with his blades. His results cannot be disputed. His methods work very well with 52100 steel. But my steel of choice is 1084, hot rolled steel. And I have found that a full edge quench all in one fell swoop works  best for  that steel. I haven't done  much work with 52100, but when  I finally get around to it I will try Ed's method of rocking for  the edge quench. Who can argue with success? I will find the best method for me  like Ed has. Through trial and error and input from other makers. That's what helps to build a well rounded knife maker. I try to  keep an open mind but it's easy to get into a rut doing the same  thing time and time again.

You had  a good question Dan.  Thanks  for stimulating my mind and getting me  to  post.

[caknives]

Thats a great question. And I have the answer! The answer is, I don't know. In theory you should be able to quench the entire blade, edge down in oil, and achieve a very similar result. As long as the edge is nonmagnetic and the spine magnetic, the results should be differential hardening. Looks like we have one more experiment to add to the list! Give it a try and let us know how it works!

[Ed Fowler]

I learned using the regulator plate from Bill Moran at his forging tutorial he taught in Laramie Wyo.
This is how I make it work for me:
Heat the lower 1/2 to of the blade to non-magnetic, leave a part of the spine non-magnetic.
Be real careful to get the ricasso area heated well, again to non-magnetic, you have to spend enough time with the torch in this area to prevent the natural heat sink due to the increased volume of steel from drawing the heat out of the thinner cutting edge before you quench.

NEVER  heat the tang to non-magnetic! A hardened tang is a major source of the source of blade failure. Bill must have repeated this information 20 times, there for I hollered at you readers using capital letters.

Once the part you want to harden is non magnetic, take a little time to even out the heat rotating the blade from side to side while heating with your torch. Do not over heat the tip, pull the torch off of the spine behind the tip or increase the distance of the tip from the flame on your strokes with the torch.

Before I start to harden the blade, I heat the oil, then I check the level of my oil and make sure that the tip end of my regulator block is a little lower than where the ricasso area of the blade will  meet the regulator block. The hardened portion of the blade will be a little lower than the oil level due to heated oil boiling away from the side of the blade.

On larger blades or blades with deep bellies I slide the regulator block back to allow me to submerge the tip a little deeper in the oil over the end of the regulator block.

Another aspect of hardening the tip is the volume of steel around the tip. A larger volume will cool more slowly than a thin tip.

I rock the blade rapidly up and down, holding the tang down in the oil a little longer than the tip. I believe the steel in the area of the tip is cooling the tip to a lesser extent while the rest of the blade is submerged in the oil. When you look at it, there is not a lot of steel out of the oil while the belly is submerged. Heat and cold radiates both ways.

Most of the blade use will be in the front third of the tip, or the belly of the blade.

All the tip has to do is get you into your work, it does not take a fully hardened tip to penetrate, as all it has to do is get you into your work, the geometry of the blade behind the tip actually does most of the work. The last thing you want is a tip that is too hard and breaks, the tip has to be pretty hard and very tough. When a tip deforms on your first test it may simply be that you did not grind enough surface material off of the cutting and point to get rid of the decarb.

When testing the tip for adequate hardness I jam it into a piece of mild steel. If the tip deforms it is too soft. I grind back and re-structure the tip to provide  adequate support and go back to engraving mild steel. I believe that when you can cut a line on the surface of mild steel with the tip of a blade without the tip breaking or deforming, you have reached the optimum balance point of performance.

This was an excellent question Dan!
I hope this answers your thoughts, if not ask for more.

[Dan Hockensmith]

Ed,
I use your method with 5160 and 52100, and believe I understand why it works with those steels. I just wanted to post a thought, and discussion prevoking question to the forum. Thank you Ed, for taking the time to post the great response. And caknives, I also have found what you say to be true of 1084, and all other simple steels I have tested.
I certainly am no metalurgist, but I have studied some, so will share my understanding of why these two different types of steels ( chromium steels  vs straight carbons steels) react the way you have both described.
After bringing straight carbon steel to non magnetic, you have to quench it to something like 1000 degees in just a few seconds for it to harden properly, as without chromium, it is considered a shallow (or fast) hardening steel. If you use the quench block, and quench the edge and ricasso first, for only a few seconds, it is too late to harden the tip, as the time of opportunity is past.
52100 and 5160, because of the chromium, is considered a deep (or slow) hardening steel, and by its nature, allows much more time to drop the temp. Hence, rocking, as you have proven, works very well with these steels. I believe it also has  something to do with the success of the 24 hour wait between quenchs. These two steel are so slow hardening, that they will even air harden to some extent.
As an example of the slow vs fast, if you heat the entire blade made of 52100 or 5160 to critical, and then just edge quench it till the heat is gone from the blade, then put it in the vise, and try to bend it, it will snap. Even with only an edge quench, the whole blade will harden. Do the same with a straight carbon blade, and the blade will bend 90 degrees and the edge will break, but the back will still be dead soft. That's why, with simple steels, clay hardening works so well. The clay insulates the steel, allowing the steel under the clay to harden just a little, depending on the thickness of the clay, giving it good strength in the spine, without being brittle.
I don't think I'm conradicting anything you have said Ed, only reinforcing it from my own perspective, and experience. Thanks for everyones responce.  Dan

[Ed Fowler]

Thank you for sharing your experience Dan. This is what we have this form for and I appreciate your thoughts. Maybe Chris and I will try that with a 52100 blade in the near future. Just to be sure I understand - edge quench the blade until cool one time then check for flex - do you let it come down to room temp. or just to the temp of the oil?

[Dan Hockensmith]

Ed,
I first experienced this while preparing for my journeymans test in 1999. I had forged a few blades, at that time of 5160. I normalized 3 times. One blade was entirely heated to 1525, edge quenched (these blades were about 9" long and 1-1/2" wide) to about 1/3 the width of the blade and kept in the oil till the blade was down to the temp of the oil, then allowed to cool to room temp before flexing. I have since tried it with 52100, with the same results.
On another note, I have had great success heating only the edge, and quenching the whole blade. For large knives, this is my prefered method. For smaller knives, the torch can't be beat. Dan

[caknives]

Ok, so even I get a little confused when I'm typeing this stuff but I always thought 5150 and 1084 were both considered "simple steels"? I'm just trying to keep these thoughts clear. Are the 10 series steels" "simple"? Need to go look some stuff up I guess. Also, If you missed it, I think Ed meant to say that he left part of the spine magnetic in his first thought at the top, not nonmagnetic. If it was nonmagnetic it would fully harden.

[Dan Hockensmith]

caknives,
Yes, 10 series steels are considered simple steels, basically meaning iron and carbon. 5160 and 52100 are chromium steels. The addition of a small amount of chromium makes a big difference in the way the steel reacts to heat treatment, and is why it is fun to experiment with.
I understand that using Ed's method of hardening leaves the spine non magnetic prior to quenching. I was just explaining my experience on the effect of chromium in the hardening of these two steels. Maybe you  misunderstood my post because I wasn't clear enough. If you heat the whole blade to critical temp, and quenched the whole blade, of course you would harden the whole blade. If you heat the whole blade, but only quench the edge, would you expect the whole blade to harden? I did this thinking that I would only harden the edge, since that is all that was quenched. However, even just quenching the edge, the whole blade hardened enough that it snapped when flexed in the vise.
If you do the same test with the simple steels, only the portion of the blade that was actually quenched will harden, and the spine will still be soft. The difference is the effect of the chromium in the steel. I think that is exactly why the chromium is in these steels. Another example would be the ball bearing. If you had a 3" ball bearing made of straight carbon steel, when you quenched it, you couldn't cool it fast enough to ensure that it would get hard all the way to the center. It would be more like case hardening. After all, you only have a second or two, to get the temp down. It would not make a good bearing. However, if you add a little chromium to the carbon steel ( 5160, 52100 ) you then have a deep hardening steel, which gives you more time to drop the temp, and the hardening to take place, and you have a 3" ball bearing that is uniformally hardened all the way to the center. Let me know your thoughts. It's just knifetalk right? 

[Ed Fowler]

One concept that I feel is significant: Wayne has been edge qun enching for as long as I have or probably more. He started a club callled the 5160 club which had a goal of pushing 5160 as far as they could.

The first thing they found when using some 5160 was that it did not develop the same as the old 5160 he was used to using. He did not know why, but was suprised by the way it hardened all the way instead of differentially.

Maybe we are comapring steels that are not equal?

Just a thought, what Wayne mentioned it to me I was impressed.!

I think that if we want to persue this, and I strongly encourage more work in this area, we got to know what we are working with.

[gmusic]

I believe the rocking method is to keep from fully quenching the entire blade.  If you tried to dip the blade into the oil far enough to quench the tip and the edge back to the riccaso you would have 75-85% of the blade in the oil.  The rocking allows you to control the depth of quench tip to ricasso.
  
Doesn’t 52100 air harden?  So if you heat the entire blade to critical quench just the edge and vise it before tempering it would break wouldn’t it.  I don’t know if 5160 air hardens.

[Dan Hockensmith]

gmusic,
I have had some 5160 that air hardened slightly.

Ed, over the years, the 5160 I have aquired at different times, has been rather inconsistent in quality. It is a valid point, that it's hard to speak with certainty about a steels quality if it is compared to the same type steel from a different melt or sorce. You are very fortunate to have a quantity of steel, of high quality, all from the same melt.

[caknives]

OK, that makes sense, thanks Dan! I didn't realize that about 5160, I have done the same thing, experiment wise with 5160 and have had the same results. I need to get some 1084 to play with ! Thanks for the info.