Knife Talk Online Forums
  Home membership Help Search Calendar Members Classifieds Treasury Store Links Gallery Media Center Login Register  
Custom Search
Pages: [1] 2   Go Down
  Send this topic  |  Print  
Author Topic: Thoughts on Bend Testing  (Read 3159 times)
0 Members and 2 Guests are viewing this topic.
ChrisAnders
Newbie
Trade Count: (0)
Sr. Member
*
Posts: 193


View Profile
« on: September 01, 2014, 06:57:29 PM »

Since this seems to be a trait of HEPK knives, I figured it best to go here. 

I was messing around in the shop/garage and did a bend test on a piece of unhardened steel I've been saving for just that purpose.  It's a rectangular bar, 1" wide by 0.18" thick.  This is in the hot rolled condition and has a hardness of around 50 on the Rockwell A scale; it's so soft it doesn't register on the C scale.  So I clamped it in the vice and bent it to about 90 degrees and notices a few things. 

First, the torque went up quickly initially, and then I could feel it start to bend.  At this point the torque went up, but much more slowly.  I measured the torque at the end of the bend and it was nearly 45 ft-lbs.

I have some pretty off the wall thoughts on bend testing, but I thought I'd see what you all thought before I started with another long winded wall of text.    

Logged
ChrisAnders
Newbie
Trade Count: (0)
Sr. Member
*
Posts: 193


View Profile
« Reply #1 on: September 03, 2014, 12:04:27 PM »

On the other end of the spectrum, a while back I discussed with Ed and others some flex testing I did on a blade I made a couple of years ago.  This blade was 5" long, 1" wide and took 5 180? flexes with no trouble.  I did not have a way to measure torque at that time.  The kicker is the blade had a hardness between 64 and 66 HRc.  It was quite thin, 0.05" or so, and when the test was over it was still straight.
Logged
Ed Fowler
Administrator
Trade Count: (1)
Hero Member
*****
Posts: 3443



View Profile WWW
« Reply #2 on: September 03, 2014, 07:12:28 PM »

It sounds like your blade did real well. When we test a blade to destruction we try to milk all the information out of it we can. it took me a long time to start using the torque wrench, wish we had thought of it earlier.

How long was the piece of soft steel?
Logged

Ed Fowler High Performance Knives
http://edfowler.com/
ChrisAnders
Newbie
Trade Count: (0)
Sr. Member
*
Posts: 193


View Profile
« Reply #3 on: September 03, 2014, 07:33:19 PM »

The soft steel was 10.25 inches from the vice to the hole where the scale was anchored.  I used a scale multiplied by the length instead of a torque wrench.  The scale maxed out at 50 lbs. so I used 50, though it was a little more.  I used a longer lever due to the scale limits. 

On the thinner blade, I think it would keep flexing as long as I cared to flex it.  Fwiw, it had the best edge retention of any blade I've tried. 

Logged
ChrisAnders
Newbie
Trade Count: (0)
Sr. Member
*
Posts: 193


View Profile
« Reply #4 on: September 05, 2014, 07:44:22 PM »

I have a few more thoughts.  They were rattling around in my head for a while now, and I found a picture in another thread to illustrate them.

http://knifetalkonline.com/smf/index.php?topic=2077.msg16442#msg16442

Check out this thread and notice how the hardened portion of the blade is nearly straight while the spine stayed bent.  Now, this means the edge is strong enough that bending it doesn't stress it much beyond it's yield point.  It is only bent a tiny bit from straight, while the rest is still where John left it, if I may call you John.  Also notice the thickness of steel at the top of the hardened portion.

I discussed this with Ed a while ago, but I wasn't able to get across what I meant with just typing.  Basically, there is a distance from the edge where the hardened portion was not stressed beyond it's yield point during the bend tests.  Between the edge and this point, the steel is acting basically as a spring.  Were you to make a cut in the blade parallel to the edge, at just this distance from the edge, the steel would spring back to straight.  This is very close to what happened with John's blade in the photos from the other thread. 
Logged
ChrisAnders
Newbie
Trade Count: (0)
Sr. Member
*
Posts: 193


View Profile
« Reply #5 on: September 06, 2014, 07:58:11 PM »

To expand on this a bit further, and tie it into the above flex testing on my fully hardened blade, I propose the following. 

There is an optimal thickness at the top of the hardened section, or at the spine in a fully hardened blade, that will allow a 90 degree flex of the hardened section (or full blade) where no permanent deformation is done to the hardened portion.  The only reason it stays bent is the soft spine holds it there.  This is what we see in John's blade in the thread above.  Once the crack ran, the hardened blade was free to straighten back out.  With what I am talking about, the edge would be able to completely return to straight after it was free of the soft spine, where John's was just a little bent. 

Now, I see something in John's very clear pictures that might confirm what I've been thinking for some time now.  But first, some background.

When flexing or bending a blade like this, the outer surfaces of the blade see the highest level of stress, either compressive or tensile.  This stress decreases as one moves to the center.  Now, for a perfectly symmetrically ground blade, with the edge in the dead center, there is zero stress on the cutting edge.  This has led me to speculate that perhaps the cracks that happen during bend/flex testing don't start on the cutting edge and go up, but start at some distance from the edge toward the spine and go both directions. 

The ductility of hardened steel is very low for the range of hardnesses used for knives (58-66 HRc).  These high hardnesses will allow a degree of ductility of maybe 5% before breaking.  When one bends a knife back and forth like many here, each bend stretches the blade a little more, particularly if the Modified Price Grind is used to localize the bending to a specific area. 

As was discussed above, below a certain thickness, the hardened portion of the blade will not actually see any permanent deformation, so ductility is a non-issue.  That thickness and it's corresponding distance from the edge toward the spine will depend on the overall geometry/cross section of the blade, the hardness of the edge, and how the blade responds to being bent, ie does all the bending happen in one localized area, or is it spread throughout the blade length.  I say this to say there is no way to predict the optimal thickness of the hardened portion of the blade without knowing or taking some detailed measurements of each blade in question, including blade length. 

Longer blades bend in a larger arc, reducing stress.  This is unless they are ground in cross section to focus the bend into a smaller region.  As an illustration, I bend tested my Kukri Machete from Cold Steel today.  The spine is 3/32" thick, 1055 steel, hardened to the low 50's HRc by my best guess.  At 55 ft-lbs it took a very slight bend (barely visible looking down the spine) and was at about 70 degrees of bend, not quite 90.

Just above the point where there is maximum thickness without stretching, the blade/edge will start to have small amounts of permanent stretching.  The thicker the blade/edge and further toward the spine one goes, the more permanent stretching will be experienced.  Eventually one runs out of hardened edge, so the stretching isn't a problem, as the steel is ductile enough to handle as much as we want to throw at it. 

However, if the hardened portion is thick enough and of low enough hardness (strength) to see permanent stretching, the ductility limit will be exceeded and a crack will form some distance away from the edge, and likely run toward the edge and toward the spine until it reaches the soft portion and then run parallel to the quench line. 

I mention all this to set up what I see in John's blade.  In the second picture, I think I see the point of the initiation of the crack.  There might be what are known as chevron marks that point to it being about half way between the top of the hardened portion and the cutting edge.  I cannot say for sure just with pictures, but if that is what they are, then it seems likely the crack started here, ran toward the edge and toward the spine at the same time, then hit the soft portion.  Once there, the crack can't continue into the much softer spine.  However, with edge quenching, particularly when the spine was not heated, there is likely a fair amount of internal stress in the area adjacent to the hardening line.  If nothing else, the edge is trying to straighten away from the bent spine, as it did for John.  This is where the crack begins to run parallel the edge. 

Now, ultimately, if the blade cross section is known or can be accurately estimated, one can determine the stress in the blade from bending based on distance from the cutting edge.  Then, it can be determined how hard (strong) to make the blade and how far up the blade to harden the edge so the hardened portion, based on it's strength (hardness), never sees a level of stress that passes the yield strength, which will allow the blade to bend until the ductility limit of the soft portion is reached, which is quite a bit, perhaps 20% or so.  There are other treatments to maximize the ductility of the spine, but that is another topic.

I will go ahead and provide my disclaimer.  I don't believe the ability of a blade to bend is an indicator of a good blade.  However, as this is Ed's forum and we are discussing his goals, I thought I might as well try to help. 

This has been a couple of years of thinking about this and rereading some discussions with Ed from Bladeforums.  It is also an interesting exercise in blade design, that may or may not provide a way to improve bendability of a blade and edge holding at the same time.  It's all theory at this point, though sound theory based on my flexing of the fully hardened blade above and the soft bar from the initial post.  I currently think that the optimum thickness for a fully hardened blade 7 inches long is about 1/16" and may be as much as 3/32".  For a shorter blade, this thickness will decrease, and all this rambling may prove impractical on say 3" - 4" blades or so.
Logged
John Silveira
Trade Count: (0)
Hero Member
*****
Posts: 860



View Profile
« Reply #6 on: September 06, 2014, 08:43:11 PM »

well - yea yea yah (= i think you're pretty spot on with where that blade of mine started to break - i'm trying to remember but i'm pretty sure it started to crack somewhere between the edge and the hardened line. But anyway i'm getting something out of the discussions - sometimes it takes talk and verbalizing to stimulate my thought processes.

Where i currently have some thought regarding development of the metal with bending and edge holding being important ( as it probably is with most people ) is ----- for example: I just finished my knife "SweePea" made out of flat stock E-Clamp. I made a test blade with the steel and got almost 3 full 180 degree bends - as mentioned before i'd be happy with one 90 degree bend (for me that's a benchmark goal). Now, SweePea did fairly well cutting edge holding as well. Was tempered at around 400 degrees . So my next test since multiple 180 degree bends is not my goal is to lower the temper 50 degrees to 350 degrees. It should yield a harder edge ( i think ) while decreasing the amount of 180 degree bends. So perhaps i'll end up with a blade that survives only One 180 degree bend instead of 3 ( which is perfectly ok with me ) but will also have a harder edge and better edge holding. For discussion sake - SweePea cut 200 times in rope then to 200 times on a log then back to rope another 200 cuts before it started loosing power.

So if i can just change temper and improve on that steel edge holding i feel is worthwile in this case with that steel. I know Ed has a method of figuring out optimum temper and i'll suppose i'll need to practice that method with the next test blade. 

Where i'm gray in my thinking is in the area of tying alot of variables together- such as - would heat treating in a different ( faster for example ) solution be the way to yield harder blade - Lets say i super quench the steel i'm talking about ( the e-clamp steel ) instead of quenching in Texaco A 18 second oil. I just can't wrap my brain around those kind of variables at the moment .. 

So lets say in my example above with the E-Clamp i do go ahead and Super Quench the next blade - would that necessarily be a better way to go about improving edge holding while also reducing the number of 180 degree flexes ( since i don't need three 180 degree bends ) ........

am i being somewhat clear about my goals.  I just don't understand enough about how steel hardens etc to be able to sit here and say through knowledge or experience how to go about getting the maximum out of this silly E-clamp steel. 

I suppose i'll just have to keep testing the steel and see how it effects the performance ROFL....   Here we go again !!!  Hahaha
Logged
Ed Fowler
Administrator
Trade Count: (1)
Hero Member
*****
Posts: 3443



View Profile WWW
« Reply #7 on: September 07, 2014, 06:53:10 AM »

This is a beautiful discussion, many variables are being discussed and this is a good thing.

One caution, we need to isolate variables, comparing different steels through testing may reveal differences in the steel, and these differences may or may not be a result of the chemistry of the steel.

Most of our work was with one pour of 52100.

In one of the photos of John's blade you can see a cross section where the sides of the hardened portion of the steel show a laminate on the outside. You can also see a pyramid extending from the hardened portion into the softer steel of the spine. The lamination's are from the same steel and formed naturally through the low temperature quenching cycles. In essence it is a bi metal blade, but all the same.

Where the tear starts may be from many variables, an occlusion or inclusion, a scratch on the surface or an isolation of some aspect of the chemistry (chemical migration) to name some common sources.

When we were testing a blade we went slow and each time we heard a 'tink' stopped and checked the blade, the first 'tink' was in the bottom 16th of an inch. We continued with the bend and another tink then more for a total of 5 tinks and the crack bifurcated at the junction of the hard and soft. Obviously there were 5 different yield points for this blade's hardened steel.

I have never seen a blade where the hard tore from the soft and returned to straight, that was dramatic and not typical of what we have seen in our 52100. I believe this is simply because your steel is different than ours and feel that is what you saw. I do not mean to imply that one is better than the other, just that they are different.

Again I congratulate you on the thoughts expressed!!!
Logged

Ed Fowler High Performance Knives
http://edfowler.com/
davidm
Member
Trade Count: (0)
Hero Member
***
Posts: 910


View Profile
« Reply #8 on: September 07, 2014, 03:08:55 PM »

Has anyone compared the differences in foot pounds of torque required to break a fully-hardened blade vs. a similar knife w/ differentially heat-treated blade, having a soft spine? 
One of the criticisms of the pronghorn used for Cliff Stamp's knife test was that the tip area deformed/bent more easily than expected when he used it to pry out of a pine board.   
Logged
Ed Fowler
Administrator
Trade Count: (1)
Hero Member
*****
Posts: 3443



View Profile WWW
« Reply #9 on: September 07, 2014, 04:02:03 PM »

The knife Cliff tested was an early pronghorn from a 3" ball bering, we have made a lot of progress since that time. If you read close the blade did very well.

We have not tried to test a fully hard blade, comparing it with a differentially hardened blade with our steel. The way we make them now you cannot deform one with your hands and I feel that is plenty good enough. Not only that if the hardened portion of the blade tears, you can straighten the blade and you still have a knife. That was our goal.

If anyone wants to do some comparisons I encourage them to go for it. This kind of testing would be interesting and it can be done with any steel you chose.
Logged

Ed Fowler High Performance Knives
http://edfowler.com/
davidm
Member
Trade Count: (0)
Hero Member
***
Posts: 910


View Profile
« Reply #10 on: September 07, 2014, 10:38:21 PM »

The last response in this thread seemed to indicate with certainty that a fully hardened blade was not as strong, nor tough as one differentially hardened. How is this fact without testing to compare the results?
http://knifetalkonline.com/smf/index.php?topic=2087.msg16833;topicseen#msg16833
Logged
Ed Fowler
Administrator
Trade Count: (1)
Hero Member
*****
Posts: 3443



View Profile WWW
« Reply #11 on: September 08, 2014, 06:59:17 AM »

I asked myself why I had never tested to see if a fully hardened blade was stronger than a differentially hardened blade?
The answer is 'because it does not really matter!" A broken blade is a broken blade.

When a fully hardened blade breaks it is essentially no longer a knife. You have a handle with no tool on the end. When a blade breaks most of the time it is merely inconvenient, you reach for another knife and go back to work.

When you are in a jackpot miles from another knife you can face serious consequences when you break a blade. I have broken blades and was in circumstances where no other knife was available, these events left a lasting impression on me. When we really need a blade and are in exciting times we can become much stronger than we are normally. I remember when I was a little kid and our neighbor had a car fall from jacks and he was pinned underneath the front end of a 56 Ford station wagon. His wife ran out and lifted the car high enough that his children and a friend could pull him out from under the car. He suffered broken ribs and was a long time healing up. There is no way she could have lifted that car without the adrenaline flowing.

I spent an entire winter apart from civilization, I had only one knife. I worried about loosing it and every time I used it I was very careful not to put too much stress on the blade in fear that I would be out of a blade. When I read about Bill Moran's work and the blades that would crack but not break I knew this was what I wanted to make.

The question: Is a fully hardened blade stronger than a blade with a soft back is meaningless to me.

The question should be: Is a broken blade better than one that will not break?

This is the motivation for our work, it does not have to be important to anyone else. If you want to make and test some blades, please do so and I would appreciate it if you post your results for us. We all make what we want to make, This is a good thing!
Logged

Ed Fowler High Performance Knives
http://edfowler.com/
davidm
Member
Trade Count: (0)
Hero Member
***
Posts: 910


View Profile
« Reply #12 on: September 08, 2014, 12:55:23 PM »

Thank you for clarifying and the examples Ed. I think i now have a better understanding of what your knives have been designed to do.  Not "sharpened pry-bars" to have the ultimate stiffness and strength like a Busse knife, but a more pliable toughness that is extremely capable to resist tearing and breaking.  Is that a fair description?   
- Sorry if my comments were aggravating.
Logged
ChrisAnders
Newbie
Trade Count: (0)
Sr. Member
*
Posts: 193


View Profile
« Reply #13 on: September 08, 2014, 05:07:52 PM »

For discussions like this, one needs to define strength and toughness very carefully.  Many times people use the two interchangeably, and I've seen threads use different meanings for each within the same discussion. 
Logged
Ed Fowler
Administrator
Trade Count: (1)
Hero Member
*****
Posts: 3443



View Profile WWW
« Reply #14 on: September 08, 2014, 07:35:42 PM »

We can add edge holding and easy to sharpen.

The last aspect of Excalibur that I wanted to develop was that the blades would not cut an honest man:
This year at Blade Show I decided to try it out, I made the statement that my knives would not cut an honest man to a few visitors to my booth. I told them that if they were totally honest and had been honorable throughout their lives the edges would not cut their skin. Many laughed, but the fact remains that even though I offered them a blade - no one was cut!

Even though no one tried do you think we have achieved Excalibur?
Logged

Ed Fowler High Performance Knives
http://edfowler.com/
Pages: [1] 2   Go Up
  Send this topic  |  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
TinyPortal v0.9.8 © Bloc
Valid XHTML 1.0! Valid CSS!