A few preliminary test...I need some help(440v,SR101,5160 related)

L

Lukers

Joined
Jul 17, 1999
Messages
571
Tonight, I did some minor testing with three knives. The knives I tested were a SRKW Bandicoot, a Spyderco Military, and one of my forged 5160 blades. I would like some input back because I had some very surprising and confounding results.

I essentially was most interested in determining the edge holding and toughness of the knife I forged. I forged it from 5160 and carried out the hardening cycle last night and the tempering cycles today. The blade is differentially hardening using the Fowler torch method. I quenched in Texaco type A preheated to just above where I could stand sticking my finger in. The temper line goes up the blade about 2/5 of the width. The knives overall blade length is 6 inches (at the start of the cutting edge to the tip). I brazed a steel guard on the blade. I have never made a knife this way before (with the guard, and a tang) I typically made knives with an integral guard or with no guard.

I did many normalizing and annealing cycles before and after grinding and heat treatment. In fact, I don't really remember how many cycles I did. I used my home-made gas forge (ask me if you want to know more about it) and ran about 3 pounds of gas to achieve a very even and controlled non magnetic each cycle. I would work on other projects while the blade was cooling in still air (normalizing) and when I shut down for the day, I would bring the blade to an even non magnetic and leave it in the shut off forge to cool over night. I was sure to stack fire bricks over the end to prolong the cooling process.

I double quenched the edge, and then triple tempered. I tempered at 390 for 1 hour, 390 for another, and then 350 for 1 and a half. Not very scientific, but it did lower the steel hardness so a file could barely bite in. I did not feel that it was wise to attach a permanent handle until after I had determined that the blade was going to hold up. When I do attach the permanent handle it will have Osage orange(Hedge, Bois d'arc, very common around here, the main reason I use it is because we have piles of fence posts 80+ years old have curred and are a beautiful brown and orange with some burl and knots) piece for the front two fingers, and a stag piece as the back two. It is best described as resembling and Ed Fowler blade, but not really(certainly no where as good!)

Okay, to the test. I ground a convex edge onto my blade with a slack belt. I finished in with a dull 400 grit. I did a preliminary toughness test by clamping the knife edge up in a vice, and pounding the nail into the edge. This did blunt the edge some but did not chip it.( I wonder if I got it too soft? Maybe 390 was too high to temper at.) Then I proceeded to go inside and polish the blunted edge out with a dull piece a sand paper and a leather strop with white roughe(sp?). The blade easily shaved hair. It actually popped hair.

Like I said I had no handle on the knife. I decided to take the handle off my Extrema Ratio T2000S (actually the handle on it is like the one from a Golum, fulcrum or Col Mischen) and I duct taped it on so it didn't wiggle. I started cutting cardboard. It was a Quaker state oil box. I cut through 100 inches against the corragations and another 275 inches with the corragations. I believe if I made more cuts through the corragations, the blade would have dulled much faster. I periodically checked the two inch section I focused the cutting on, every 30- 40 inches of cardboard. After the total 375 inches, the area used could barely scrape hair. I stopped and thought "hmmmm...now what?" I decided to see how easily the edge would come back. I used a longer piece of the cardboard I cut as a strop and made 5 or 6 strops per side. The edge would now easily shave my wet arm hair(darn humidity), but it was not as sharp as before the cutting test.

I decided to compare it to the 440v blade of my Spyderco Military. I brought the Military to a very similar hair poppin edge (as close as I could tell) using the same dull sand paper and the same leather strop. I started cutting. I cut 100inches of the cardboard from the same box against the corragations and cut another 150 inches with the corragations before it would just barely scrape hair in the two inch section. Frankly, I was shocked. I still am. I remember reading about the CPM steels(and specifically 440V) in Wayne Goddards book "The wonder of knife making". I was expecting the 440V to last atleast 3-4 times as long as the 5160 blade. At this point I am holding off a conclusion because I simply do not believe it. To me, it is more believeable that I am unknowning biased then to think that my 5160 blade held an edge longer than the Spyderco in 440V simply because of some subconscious superiority complex(hey it could happen). I can't determine what the deal is. I think I will just have to start over and re-do the test over. I may have to try rope cutting or something different.

I next wanted to see the relative level of damage in the Swamp Rat blade compared to my forged 5160. The Bandicoot has a more substantial edge thickness than my blade, and is twice as narrow(width wise) as my knife. I took the same framing nail I used on my knife and used the same technique to cut into the nail. Overall I think that the Bandicoot held up better than my blade. It could handle harder and deep cuts with about the same damage. The only problem I had with it was that the edge chipped out. A section .140" long by .015-.020" deep chipped out of the edge. I was surprised. I actually found the piece that chipped out, but lost it a little later. I am not sure why it chipped out. Again, I am not ruling it out as the blades fault. I simply don't know.

I was fully expecting my 5160 knife to get spanked. I was prepared to have it loose its edge and chip and ding out. While it did loose its edge and get a blunted edge from the nail, it did surpase my expectations(I guess this means I have to put a permanent handle on) However, even though it did what I wanted it to, I will not say after one simple round of tests that it surpased the two others. I think there may be too many variables I am not accounting for or something. It was my goal that my knife combine aspects of the Military and Bandicoot, as a feel they are both a very high benchmark of the production arena for a cutter and a tough little thick edged bruser.

Maybe someone could shed a little light on this for me? What can I do differently to push this test further and eliminate inherent flaws?

Thanks
Luke
 
I can believe the results. I don't know much about the Swamp Rat knives, but I know the 440V Military very well. I've even run cardboard cutting tests with the Military, directly against convex ground Marbles knives. The thin convex grind of the Marbles made a whopping difference, where my Military had a relatively thick flat grind. The contest wasn’t even close. I truly wasn’t happy with the performance of my Military until I thinned and convex ground the edge.

My bet is that you’d have a hard time testing your knife against other knives, in order to actually test the steel and not the grind, without using similarly ground knives. The grind is a big deal.

It’s sounds like you’ve got one heck of a knife going there!!!
 
Regarding the edge chipping on the Bandicoot, I have used two Swamp Rat knives, Camp Tramp and Howling Rat, and have seen similar behavior. When in contact with hard objects they will chip before denting significantly, 52100 is not as tough or ductile as 5160. You might want to compare both knives on cardboard to see if the Bandicoot has better edge retention and a flex test.

In regards to the Military, Spyderco is running the Military in 440V (S60V) very soft, ~55 HRC. Assuming your knife is much harder, and it likely is about 60 HRC, it is no surprise that your knife has better edge retention becuase early blunting on carboard and similar materials (loss of shaving) is mainly a rolling effect which is critically dependent on strength which is strongly dependent on hardness.

If you want to eliminate bias, in the middle of the cardboard trial, leave the room and have a friend make a certain amount of cuts, but not tell you how many they made. Now finish the work yourself until you have reached the stopping criteria. Resharpen the blade and repeat this a few times, asking your friend to do a random number of cuts (the second hand on a watch can be used as a random number generator).

Now once all the work has been done, get the numbers from your friend and see how consistent your work was. If your numbers run all over the place with no coherent grouping it is very likely that you are using far too much personal judgement and intruducing a bias in your result. However if you get a very nice grouping you can feel confident that you are not letting your own desires influence the results.

I do things like this periodically to keep myself honest. There are lots of other ways to do the same thing. For example when you are testing different heat treatments, have a friend mark the blades with a number, but not tell you which one was which. You then do all the tests blind and only at the end find out which one had the "advanced" heat treatment.

As a side note, when making posts such as the above, it would be of benefit to the reader to make a small summary comparing the results such as :

5160 Custom : 100" against corragations, 275" with

S60V Military : 100" against corragations, 150" with

-Cliff
 
I'd agree with comments from both of the esteemed posters above. The grind makes a lot of difference, as does the edge itself. At the very least I'd make sure the edges on all tested knives are as similar as possible, even if I had to put a convex edge on the Military (I actually slapped a convex edge on my endura a few months ago for the same reason, though doing that to S60V is bound to be a drag). But even then, the grind above the edge can matter.

Also agree strongly with Cliff's statements, especially regarding the Military's relatively low hardness. 2-3 Rc points definitely becomes noticeable as far as denting and rolling go. So it's possible that your test was completely valid, and you're just not seeing where it's leading ... not to 5160 being more wear resistant than S60V, but more a general statement that with cutting materials that offer some resistance, edge strength may be more important than wear resistance. To help validate that conclusion, test your military against your 5160 knife in a test that's more targeted strictly at wear resistance and abrasion (ideas anyone? carpeting? hemp rope held looped in one hand?). If the Military proves more wear resistant than your knife for abrasive cutting, but your knife does better for more forceful cutting, we have the beginnings of a theory on cutting and how wear resistance and strength interplay (in fact, you come to the theory that Cliff's been hammering out piece-by-piece for years).

Joe
 
Carpet cutting is fairly abrasive among other things. You can test for extent of rolling by seeing how the edge responds to a steel. The extent of wear can be tested by noting the need for abrasive honing to restore sharpness. Count the number of passes in each case.

You might also want to look at sharpness beyond just stopping to shave. Try cutting until the knives tear the cardboard for example. The longer you use a knife the more that wear resistance comes into play. As well try out different grit finishes.

S60V is a heavily carbided steel, its performance may be enhanced more than the 5160 knife if the edges are left very coarse. Try leaving them both with the finish from a 100 grit AO belt (or similar).

You can also try out edge retention for chopping or splitting, this is very critical on hardness and ductility and pretty much ignores wear resistance.

-Cliff
 
Considering that carbides play a very important role in wear resistance and S60V has far more carbon (almost 4 times as much) than 5160 ,you were testing grind more than anything else. To perform a valid test you have to understand exactly what the variables are and exactly what you are testing.
 
My guess is similar to Cliff's, your edge is a lot harder than the 440v. What surprises me is that your much harder edge did not chip. 5160 is known to be tough, but I didn't know that it would display that toughness at RC60 (I didn't realize that it would be useable at that hardness, it doesn't have very high carbon). I would have guessed that it was around RC58. One way to try and quantify would be to do a comparitive file test on the SRKW blade verses your blade. I assume that SRKW quotes a Rockwell number. If you use a new file you should be able to feel which blade causes more of a skating feeling when you file. See if your edge is harder or softer than the SRKW.
 
In regards to chipping, edge geometry, especially at the micro bevel stage is critical.

Highly convex edges, ones with a lot of curvature are very durable in regards to chipping not only because they are more obtuse at the very edge, but because the curvature will act to deflect stress away from the very edge.

Flat beveled edge tend to do the opposite and instead focus lateral strains across the edge. So when making a steel comparison on toughness in that respect, look at the edge geometry very carefully.

Regarding 5160 at high hardness levels. One of the toughest blades I have seen was a villager khukuri from HI. The steel was very hard, very difficult to file, and yet very chip resistant. If I had the time back again I would definately had that one HRC tested.

-Cliff
 
Lukers,

Since you are a knifemaker, maybe you could make yourself a Military out of 5160 and test it against the 440V Military. Of course there is no need to make a folder, you could just follow the Military's blade geometry and make a handle with scales for the same feel.

You would probably have to run the tests a few times to see how consistent your results are.




- Frank
 
Lukers:
Thanks for the interest in cut and testing your results.

Next blade try this, temper the first three times at 350, then try the steel or brass rod test, flex the edge using a drawing motion, if the edge chips, then temper three times ten degrees hotter and try the edge flex again. When it quits chipping is the time to test cut. Tempering temp may vary with diffent batches of steel. Be sure to grind out all the portion of the edge that chiped before you temper and test again, and naturally before you finish the knife blade. 4 to six flexes with no chip is a good indicator of a blade that won't chip in use.
Good Luck!
 
The brass rod test is an unspecified bend test. I have take the same knife and had it pass the brass rod test, or fail it by staying bent or breaking depending on how hard it was pushed on. Aside from how much force is applied, the angle of the edge is also a major influence as is the diameter of the rod.

For the test to allow inference of steel properties you would want similar edge angles and the same rod size for all tests, and as well note the onset of bending (how much force was applied), width of the plastic region, and the yield point. This would give measures of strength and ductility.

Ref :

http://www.bladeforums.com/forums/showthread.php?s=&threadid=178558

Note that this test also gives no information on impact toughness which is of critical importance in various aspects of durability. Though ductility and impact toughness are highly correlated, they are not identical and you would want to look at both for a complete picture of performance.

-Cliff
 
I thank you all for the feed back on ways to test stuff and glean more from each session. I did compare my knifes edge hardness to that of the Bandicoot and an Extrema Ratio T2000S. I believe the file bit into my edge just ever so slightly more that the SR and ER. The ER is supposed to be 58 RC and the SR is supposed to be about the same. The ER and SR seemed to "skate"the file the same.

I will try making a knife of identical edge geometry to the Military. One thing I do like about the Military is that the blade width is seemingly proportional to the thickness(I think that means a distil taper). I have put a slight convex edge on the Military, but it probably wouldn't be considered truly convex.

Ed,
Thank you for the encouraging words. I admire your work(even though I have never seen one in person) and believe that your knives have the safest and most beautiful design characteristics. Are you saying the hottest tempering temp will set the hardness level? What would the next two cycles do to grain structure, performance, sharpenability? Would there be any advantage to numerous tempering cycles(say instead of 3, maybe 6 or 12) Since the tempering temp seems to approach the martensitic transformation range, is it the goal to bring it up past it and then down again to get a more complete conversion?

I brazed the guard on the knife. Do you believe that the structural integrity of the blade is effected by this process? Is there a better way to securely attach guards? I learned a valuable lesson that I should not use 3/32 rod for such a small task. It really puts too much on the guard and blade, which I had to melt out later and then do a lot more finish work. I attached the guard before the hardening cycle. Is this the way you would do it? Do you think if I'd etch the blade and post a picture, you would be able to tell me something about the grain structure?Do you have your second book available at this time? Thank you very much!


Cliff,
I am not sure about the brass rod test. I tried it with a sharpening ceramic, but must have held it too flat because it simply formed a wire edge. I do understand what you are saying about the diameter of the rod being important. The smaller diameter, the more force per area. But it does make sense if the test were standardized, that a blade that is tough and holds a good edge should be able to flex over a certain area x to indicate the compromise of toughness and edge hardness.
I did several more tests with cutting into framing nails with my test blade. I did it several places along the edge and the damage seemed to be uniform with each other. The damage was very small though. Well before I didn't want to push my luck, but I decided what the heck. The edge would not chip or bend, but it would blunt. These nails are fairly hard. It is difficult to cut them with a file. I don't know their hardness but I would guess it to be in the 40s RC.
 
Lukers :

But it does make sense if the test were standardized, that a blade that is tough and holds a good edge should be able to flex over a certain area x to indicate the compromise of toughness and edge hardness.
Click to expand...

It depends what you mean by standardized, if you mean measuring the force applied then yes.

All steels have an elastic region, and thus any steel can pass or fail the test. If you want to compare steels you would compare the strength, and the width of the elastic and plastic regions.

The brass rod test, as commonly described, is an unspecificed bend test. It basically reads, bend a knife, if it returns to true it passes, if it bend and stays bent, or bends and cracks it fails.

Any steel can have any of these behaviors depending on how far you chose to bend it. With some steels the plastic region might be very narrow, so care would need to be taken to find the set point, ATS-34 for example.

And note again - this test is not influenced at all by impact toughness which is the critical factor in many kinds of chipping. All it tests is the extent of the elastic region.

Note as well, that the elastic region is actually constantly exceeded in use. As soon as you cut a piece of paper for example the edge has taken a set, a small one, yes, but a set none the less.

Cut a little cardboard, and then look straight down at the edge under magnification (10-20x) and you can easily see how distorted the edge has become. Thus the critical point here is the ductility, which the brass rod tests also ignores as blades that pass the test never have their ductility tested.

-Cliff
 
Lukers:
The edge flex is a skill, learned by experience, the more you do the better you will read the resultrs.

I silver solder my guards, using low temp silver solder, about 420F, the ricasso is a heat sink that allows me to heat the joint without influencing the improtant part of the blade. I have thought about brazing the guards and decided against it due to the high temps required. If I did braze I would probably follow it with three 800 degree 'blade smith anneals' then harden and temper.

I do not see the benefit of lowering the tempering temp each time. Any experience of a higher than needed tempering heat will set the table for that lever of hardness. I temper, holding the heat for 2 hours and let it go up and cool to room temp slowly each time. Each cycle takes at least 7 hours. Then test for flex, once flex is established I test for cut. Anyone can make a hard blade cut, the functional balance between cut and flex is part of the nature of the quest for Excalibur.

Richtig obviously had a very effective method of testing cut and tough, I have some experiments I want to do to check it out (a hypothetical construct at this time). I will report what I find if they prove valid.

My new Book, Knife Talk II is out and available from Krause or an autographed copy from me. From me it costs $20.00 + $3.50 postage. Thanks for asking and most importantly thanks for joining in testing and seeking better knives.
 
You must log in or register to reply here.
Back
Top