Another Quench Question

BD Blades

Well-Known Member
I was reading a post about Ht and someone mentioned using the wrong quench medium would cool the blade improperly and it would skate a file all day, but not hold an edge.

I have gone back and searched and could not find that thread. Could someone explain this further as I think I have done this before and it has been bugging me for a whole week trying to find more info on the details of this accuring.

Thanks for your help.
 
I was reading a post about Ht and someone mentioned using the wrong quench medium would cool the blade improperly and it would skate a file all day, but not hold an edge.

I have gone back and searched and could not find that thread. Could someone explain this further as I think I have done this before and it has been bugging me for a whole week trying to find more info on the details of this accuring.

Thanks for your help.

I would hope the discussion you remember reading out there was more about the pitfalls of seeing one test as the definitive final word on something rather than misleading folks with any absolutes of its own. Can a poorly matched quenchant lead to drop in edge holding? Yes. Is that a certainty? No. Is the file test worthless in light of these circumstances? No.

But to the very heart of the issue- is the file test fool proof or is there still room for improvement if a blade skates the file? No and yes. Should you have less than successful hardening operation and have fine pearlite colonies (the soft stuff) intermingled in the martensite (the hard stuff), the file test is macro in nature and will only give you the overall picture by skating over the martensite and not detecting the soft pearlite specks.

Think of a ball of soft clay filled with shard of glass, a file will skate over the glass and give you the idea that the whole ball is as hard as glass, this describes the file test. But if you turn the file around and remove to handle to plunge the pointed tang into the clay ball it will simply push the glass aside and tell you that the ball is as soft as clay, this describes the Rockwell hardness test. The truth is somewhere in the middle but we will never find it if we believe that any one test can tell us all we need to know.

I believe the conversation may have been in attempt to help people improve their results by not falling victim to the concept that there quench was working just fine simply because the results skated a file. There is always room for improvement.
 
The file test is a scratch hardness test and the Rockwell a penetration hardness test. Two different things. Both can be useful, but neither will tell you everything.
 
What Kevin said plus be aware that statements like you sited will depend on the steel, the quenchant plus a few other factors, like austinizing temperature. The art of making and heat treating knife blades is full of trade offs and exceptions and answers frequently depend on the exact set of circumstances involved.

Doug
 
Just trying to point out a fundamental difference.

I do think that a Rockwell test has less variables and acts as it's own control. However, if the pit falls of the file test are understood and taken into account, it can be useful. With any test, you have to have some sort of control or something to compare to. So, simply picking up any old dull file off the bench and seeing if it skates and/or if the steel “feels“ hard, really won't tell you much at all. It has to be much more controlled than that, to be of any use at all.

I don't think the file test will every completely eliminate all it's variables and there will always be some element of skill, experience and intuition involved.
 
Last edited:
Thanks very much for all the info. I do understand better, that was driving me crazy. I just use the file as a guide. The main things I go by are edge flex and cutting test. I usually cut 3/4" hemp rope. Those aren't all inclusive either, but the main thing I expect is the knife to cut and hold an edge. I also use the file test as a guide I have had a file bite into a blade I thought was hard a few times.

Thanks again guys
 
Thanks very much for all the info. I do understand better, that was driving me crazy. I just use the file as a guide. The main things I go by are edge flex and cutting test. I usually cut 3/4" hemp rope. Those aren't all inclusive either, but the main thing I expect is the knife to cut and hold an edge. I also use the file test as a guide I have had a file bite into a blade I thought was hard a few times.

Thanks again guys

Reliable and accurate tests are geared to eliminate variables and test a very narrow group or a single specific property. The single largest pitfall in knifemaking these days is in the accurate interpretation of the results of any given tests, it is one thing to do a test, it is another to know what it is, or is not, telling you. Files and Rockwell testers both measure hardness but hardness come on a couple of forms, in fact hardness is more a concept we use to represent different types of “strength”. Rockwell measures deformation much like Brinell, Vickers and others ,the file is more along the line of the Mohs scale often used in minerals.

The edge “flex”, like any flex testing, is most accurately used to measure consistency in thickness and geometry, not so much heat treatment. There is a claim passed around modern knifemaking that it was a test used in old German cutlery facilities, while this is true what is left out of that tale is what part of the production it was testing. The guy responsible for the edge flex test stood behind the guy who was grinding the blades to insure his grind consistency.

I do use the file test often, but always in conjunction with Rockwell to get a more complete picture. And you touched on the time when the file yields very good information- when it easily bites into the steel. There is no mistaking that. If the file bites the steel, the steel is at best softer than the file. The only time this may be a false reading is if you test too soon after the quench and the blade is still in the process of hardening, but a follow up test a little later will reveal this.

I think it is a difference between “pass/ fail” type tests, such as the file and the edge flex, and tests that give us solid numbers on a scale, like Rockwell. The former can only tell us something if the blade fails, they don’t give us degrees of success by which we can even further improve our blades. Thus many folks are never aware of the untapped potential and gains they could make if they weren’t being told that it was “good enough” by limited or misleading testing.

P.S. BD Blade, and others who start other threads with good questions , I am sure you already realize these things, so don’t take my continued expansion of the topic as an indication that I feel the need to lecture you. But the reason we do this on a forum instead of with e-mail is for the benefit of all those who will come here and read these threads looking for answers to their questions. I constantly am contacted by people who have found things I wrote years ago and have new questions. Each of these threads is also for all of them as well.
 
Last edited:
Some of the variables are, the file itself, type of steel being tested or it's degree of abrasion resistance, amount of surface area of contact, and the amount of pressure or force applied. The condition of the file also plays a major role. A dull file will "skate" at much lower hardnesses that a sharp file. Most of the time folks tend to use the flat of the file for testing. However, at higher hardnesses the sharp corner of a file will usually be more tell tale and you can actually see whether or not the file is cutting by the presence or absence of a small depression. I think using visual clues whenever possible is also helpful. I also usually lay a sheet of white paper down under the blade when testing so I can see if any particles are coming off and to what degree. In many cases the file will feel like it's skating, but when you see particles coming off you know otherwise.

Another very helpful thing is to have a control group of samples of known steels of known hardnesses to compare to. These could be coupons or other tools,... blades, bits etc. There are sets of Rockwell hardness testing files, but the problem is they get dull with use. However, they could be used more effectively as your control group for similar types of steels, by comparing them with a separate testing file against the blade in question.
 
Last edited:
Thanks guys,

I do hope other will benifit from this post as I have from so many of the posts on here.

I do have access to a rockwell tester at work. I can't get every blade tested, but do have some from time to time tested. The knife in question as far as this post goes. (I have had this problem about 3 times) Passed the file test for what is worth very well. It also tested at R 58. It was a 3" 5160 blade, and would get really sharp for 5160 I thought. I could only get 45 cuts on 3/4" hemp rope from Lowes. The blade was thin, and I even thinned it more, sharpened it a couple of different ways and angles, but 45 cuts is all I could get.

I realize there are many things involved in a blade cutting well. The steel, HT, geometry, shapening and so on, but IMHO I think this time it may have been the HT/quench in it not hardening or a having the mostly pearlite. I have only made about 15 or so blades, but was expecting to get around 100+ cuts on the rope. So far my best cutting blade has done 159 cuts on 3/4" rope before it would barley cut rope at all. I did have one other larger knife make 125 cuts and then chop through a 2x4 and still shave hir from your arm. There was some dragging when it shaved though. I have also tried the standing water bottle test, but can only get 1 and a half bottles cut. The knives I make are really not the proper size and shape for that though. There is also probably some techniqe involved as well.

There are probably also other ways to test that I am not aware of, but these at least give me some ideas. The primary concern is cutting and edge holding. I am open to any other testing methods you guys recommend.
 
You can also try cutting soft steel wire or a thin brass rod by driving it into or through by striking the spine with a mallet. It shouldn't leave anything more than just the slightest dent in the edge. If it dents and the edge of the dent looks rolled over then the steel is too soft. If it's chipped looking then it's too hard.

5160 is a rather deep hardening steel so it is less likely that you are clipping the nose of the IT curve and getting pearlite mixed with martensite than if you were using something like 1095. There is also a possibility that you got the edge of the blade too hot when you were grinding it after heat treatment and over drew the temper in the edge. Remember to grind a little and cool in water often. Also hold the blade with your bare hands so that you can feel the blade getting hot. Also if you get a blue patina develop on the edge or, worse yet, it starts to glow then you could have ruined the temper on the edge and you need to repeat the heat treatment.

Doug
 
Thanks Doug, I will have to try the wire test, sounds like a good one. I guess I also need to bend and or break one. Probably should have already done that, but haven't. I don't currently have a vice mounted up strong enough to do that test, but can probably solve that easy enough.

I do keep a water bucket on an old bar stool right next to my grinder and usually make no more than 2 passes when grinding after HT. I always grind bare handed.

I really apprciate all the help from you guys. I hope to someday be the one giving the help.
 
Interesting thread...

"Not everything that can be counted counts, and not everything that counts can be counted."
- Albert Einstein
 
Back
Top