Knife sharpness - rope cutting - edge testing machine

Joined
Feb 23, 2002
Messages
1,088
I have been thinking about how I could come up with a simple machine that you could mount any knife to and a roll of any size rope.

Then turn it on and watch it cut the rope, count the cuts and auto feed the rope.

You could adjust the force that the knife comes down on the rope.
You could adjust how tight the rope is pulled befor it's cut.
You could adjust it for a slice cut or a push cut.

Once these adjustments are made one could test differant knives with differant edges, differant types of steels, etc....


All these idea's have just been in my head and on paper, but I'm about ready to start building.

Does this interest any one else?

I would be interested in hearing your thoughts on this idea.
 
Go for it! From other threads, you probably already know how I feel about the need for objective standards in the knife testing field. Anything you could come up with that would yield reasonably accurate results, would be a quantum leap forward. Keep us posted.;)
 
Speaking from a purely business end of things, do you think there would be much of a market for what seems to be a rather complicated machine that tests a relatively simple (albeit time consuming, and somewhat subjective) process? I certainly wouldn’t want to discourage you, and if the price were right I would probably be interested in something like that. However, I how big of a market you envision for this.
 
This is something I would use myself for testing.
Then a I could post results from differant knives.

I was not thinking about selling anything, but my idea's are always for sale :)
 
Pardon me while I find a towel for that egg. :o

I leapt to the conclusion that you were thinking of a product to sell. I suppose you should just ignore everything I said in my first post.

:D


Well if your idea works like you are anticipating and the cost wasn’t prohibitive it seems like a possibly marketable product. If it was affordable for the “average joe” I think lots of people might consider snagging one.
 
If you are able to buy something just for yourself and testing different blades, I´m sure everyone here would be very pleased and thankful.

There is just one thing one must keep in mind: the handle shape is very important for cutting comfort and efficiency when used by a human. These subjective criteria can hardly be tested and if, then not with a machine.

But for simply comparing different steels with a different heat treatment, various edge profiles and blade shapes, this is an extremely good idea.
There is this rust-resistance-test where salt water is sprayed on the blades until they show rust. As soon as any corrosion is detected, the test is stopped and the longer the knife lasted, the better the corrosion resistance of this particular blade.
The only problem with this testing method: it´s very rarely found. I only heard about very few blades which took part in that contest. If many knives are tested, one can compare them easily, but if you just performed this on a few knives, the results are not very useful IMO.
 
Pahl,

This is a good thing you are thinking about. Following many of the opinions expressed by the thread you started on Cliff S., some objective testing would be good.

Being a research-oriented guy myself, I would follow up on what Quiet Storm said with just one point. The more factors you can hold constant while testing the cutting abilities of a knife, the better, IMO. For example, in the testing of S30V done by Phil Wilson (Blade July, August 02), he constructs knives of the same blade style, length, edge geometry to test different steels to S30V.

What would be neat if you make this machine, is to be able to test the same knife in different steels. Maybe some maker would want to participate in the testing.

Your findings will be appreciated by many of us. Since so many variables are involved in actual knife usage (ergonomics, strength, swing styles, cutting angles, material types, etc.) knife testing always is controversial. Hang in there!
 
Very interesting web site.
I won't waste anymore of my time on any idea's I had.
As always someone else is way ahead of me :)
 
Pahl, it was not my intent to discourage you. I just thought the CATRA machine would interest you, and maybe give you some more ideas. If you want to build a tester, please do, I would be interested in your results. - Frank.
 
Test those knives, and give us un-biased reports. You are separate from the knife industry therefore have nothing to prove, exept for honest test that takes out the human factor.
-Scott
 
If you want to test the wear resistance, strength or other properties there are already well known standard tests. If you want to test how these properties influence cutting ability, edge retention and so on, (which is a good thing), you have to be very careful not to get so distracted making everything controlled that you lose sight of what you intended to do. The Catra machine very well quantifies such things as the force used in a cut, so you can see the effect of various edge geometries. However when Buck tried one of the IonFusion blades it scored insane on the Catra machine, but it doesn't do nearly as well when people use it. People don't hold blades in the same manner nor can they do cuts in as controlled a fashion as a machine.

If you want to understand how blades respond to cutting by people then you have to look at results from people doing cutting. In order to gain the most information, don't remove this random element as this removes one of the critical sources of the information. What you need to do is find some way to estimate the forces used for cutting ability, and some way to measure the stopping point for edge retention. Once you have this done, even in a rough manner, start doing some work and seeing just how blades blunt and how geometry has an influence. You can always upgrade your method as you are doing the work when you want to look at the finer details.

-Cliff
 
Originally posted by Cliff Stamp
However when Buck tried one of the IonFusion blades it scored insane on the Catra machine, but it doesn't do nearly as well when people use it. People don't hold blades in the same manner nor can they do cuts in as controlled a fashion as a machine.

If you want to understand how blades respond to cutting by people then you have to look at results from people doing cutting. -Cliff

This is one of the points that's frustrated me a bit in the "testing" discussions. It's very easy to end up with a machine that does things very precisely, but doesn't subject the knife to the same forces a person does. So you end up with results that are objective, but irrelevant, and none of the machine-touters seems to want to address this. In my mind, a machine-based test will have interesting results, but without real live hands-on tests, you really can't learn a whole lot. Yes, tests by real people have some slop in them, so just make things as accurate as you can, and don't get carried away carrying things out to the n-th significant digit.

As an absolute minimum, I want to see hands-on tests as a basic verification of the machine results; but for real-life relevance, the hands-on tests have to be given a lot of weight.

Joe
 
Joe Talmadge :

... don't get carried away carrying things out to the n-th significant digit.

Yes, from a basic perspective, is it even meaningful to measure performance to such degree beyond what people can observe naturally? Of course not in the context of a review intended for use by people. Just think about the goal and this is obvious.

By hand you can easily see differences of ~10% with a little care and repetition to smooth out the random effects. About 5% takes a lot more time, and going beyond that starts to take a lot more work (it isn't a linear process). Of course more power to you if you want to work in that much detail.

Pahl, let me know if there is any way I can be of help in any case.

-Cliff
 
Here are a few thoughts on this:

A machine to test the edge retention on rope cutting, will only tell you how good the edge retention is, when cutting rope, done by a (or even THIS) machine. This is a very restrictive view of the different parameters that make a good knife for use by a human.

It is a very sad view of "science" or "scientifical experiment" to consider that the observer is not part of the observed system. Of course he is:

The very experience you want to design in the beginning shows that you are more interested in edge holding than in sharpenability, edge efficiency (ability to penetrate), or even ease of use of a knife. The results you will get will be pure numbers on one parameter, and though true and factual for this parameter in this conditions, will lead to as many interpretations as you'll get readers because it hardly applies to real life. Where is the objectivity in it? Are knives designed to cut only rope, as it seems THE test to do ? Will a knife that cuts rope well, perform on beef, horn, wood, plastic? Talking about objectivity, has someone ever proved that edge-holding on rope can be extrapolated to represent edge-holding in general? Which edge holding are you talking about, slicing or push cutting? Is slicing edge-holding correlated to push-cutting edge-holding? if it is, who prooved this ? How do you ensure the sharpness state is the same at start?

On all these questions we have empiric answers( read Joe's FAQS for a start), and I find them statisfying. Let's not get knives for high precision mechanics, where 1% difference in performance has a meaning. As Cliff said the goal (for most, there are some collectors and devots too) is to use, and therefore what cannot be perceived is not important.

You could say that you could design tests for all or most parameters. I'll reply that you'll never know all parameters, and that if you take the most importants and rate them, you still do not know if a knife that scores great for all tests is a good knife.

In this the human factor IS the key, and surely like our tastes are different, appreciations changes from human to human.

Testing a knife is to me like testing a car. Sure you can make numerous crash tests, and they will tell me how (hopefully) secure it is in some (controlled) extreme conditions. But only a driver can tell if it behaves like a tank or a ferrari (which does not mean that there is not receipe to achieve the final desired result). And sure you'll get different appreciations from a usual limousine driver than from a usual sports car driver. The only thing to do there is to try to understand if the tester has the same taste and habits than you. But if the same driver tests another car, he'd be able to tell you which one is the more powerful, without reading the specs.

Coming to comparative review, then, rather than designing machines that none will have access to (and this is generally the stopping point to the spreading of standard mesurements), why don't we choose a cheap set of "reference" knives we can all own and compare new aquisitions to?

A last addition, I really find edge-holding to be a fashion and hype parameter which allows to sell knives which are often otherwise badly designed, a marketing argument targeted at people that ignore how to/cannot sharpen. Anyway, with steel, as everything has a price, and is a compromise, you generally pay the bill when the need to sharpen arises...

No intent to walk on any feet or aggress anybody, I just express my thoughts, and they are different. I have STRONG doubts when anybody starts talking about objectivity like if it was a switch, on or off.
 
a shear-cut tester. a rotating wood or steel rod is wound up with dry or moist newspaper till it's 10" in diameter. a spring mounting for the knife will press its edge against the paper roll. when the roll is turned, the mounting will mimic the human hand doing a continous draw cut around the roll. depending on the time lapsed, number or turns and pressure of the spring, measure how deep the blade had cut into the roll.
 
I was following the other thread and couldn't even finish it. Scientific vs subjective testing argument ...

My opinion: Objectivity as in Newtonian concept is over. Now we are in relativity (Einstein) concept and quantum era. Fritjof Capra (the Tao of Physics) and many modern sub atomic physics professor have stated that pure objectivity doesn't exist. The observer is part of the test and the two is inseparable.

Having said that, I am not saying objectivity in Newtonian concept is wrong. There are situation that best served by using the concept (e. g. aeroplane load calculation) but I believe (at least to me) the majority of us who read knife review / testing would be interested not in the scientific test but to know how 'good' a knife (Robert M. Pirsig: Zen and the Art of Motorcycle Maintenance) is. Like Singularity and Cliff said: the goal is user/owner satisfaction.

Now in this case, my suggestion is:
1. Use the machine to test a set of standard production knife that is affordable for majority of people.
2. Continue the test manual and compare the result (something like Cliff does)
3. Publish the result

We then can have these knives as reference and when there is a new model available, just follow step 1-3 and we (the audience) can have a feel of how 'good' it is in comparative manner.

However, final check still should be with the user: handle a knife and feel it before deciding to buy. No amount of review reading can compare to how you feel about it. Example: I never know Spydie Police SS model can be so big and heavy before I saw and hold one. While it is an excellent knife, it is not suitable to me for EDC. People's review of this knife made me want to have one, but ...
Despite the better approach above, sometime I do gamble a bit: I ordered a CRKT Large Point Guard without handling one first, but I can afford spending the money (and give the knife away later if I don't like it) on the little 'gamble'. This is where review has more importance. For other cases, the review will make me want to check on a knife.

As the old saying goes: you can't get drunk reading the wine review ...
 
Singularity :

The very experience you want to design in the beginning shows that you are more interested in edge holding than in sharpenability, edge efficiency (ability to penetrate), or even ease of use of a knife.

It could simply be the first thing he intended to look at. The first thing that I do when I get a knife is measure its NIB sharpness, however this is last on my list of what makes a high performance knife.

Are knives designed to cut only rope, as it seems THE test to do ?

It is done mainly because it is very easy to get a lot of rope cheaply which is very consistent. It has a lot of great characteristics though for looking at knife performance as it is both resistant to being cut into and resistant to being deformed so it allows you to look at both aspects of cutting ability (rupture and binding), as well as push cutting vs slicing. Cardboard is very good as well, I don't do a lot of it simply because I don't have access to it.

... will a knife that cuts rope well, perform on beef, horn, wood, plastic?

Unless the edge strength is exceeded and it breaks apart, the basic geometrical effects are the same for all materials so yes. What changes are the compressive strengths and hardness of the material which adjust the influence ratio of the rupture resistance and deformation (how hard it is to start the cut, and how hard it is to continue it as you have to wedge the material apart). For example cheese is very soft and easy to cut into, but wedges tightly onto a knife. So a thin blunt knife can slice cheese better than a thick sharp one, and the opposite will be seen on rope.

Since most rope cutting is done on narrow cord, the primary grind doesn't have a significant effect (Busse is one of the few people that cuts full one inch rope). A blade with a very shallow sabre hollow grind and a thin edge, can cut thin rope very well, better than a SAK say, but not perform nearly as well when slicing pine, as you get 2-4" of penetration there and a thick primary grind will wedge badly, especially the "t" shaped hollow grinds. Of course they are the same basic principles as what is found in the rope cutting, so once you do a lot of it you can extrapolate it to other mediums and predict the performance based on how hard they are to cut and how badly they wedge.

You can also adjust the depth of the cut to focus the performance on the aspects you want to examine. For example thread looks at sharpness, 1/4" poly at sharpness and edge angle, 3/8" hemp at sharpness, edge angle and edge thickness, 1" hemp at sharpness, edge angle and width and primary grind characteristics. I use wood instead of 1" hemp as it is cheaper (I have a lot of it at hand). Lots of other materials should be included as well of course to examine the effect of different properties, for example cut thick vegetables. They are very soft so they are easy to start a cut, but are very elastic and stiff so exert a lot of wedging force on a blade.

... edge-holding on rope can be extrapolated to represent edge-holding in general?

Edges blunt by rolling , wear, fracture, indentation and corrosion. Rope is mainly rolling and then wear, with light indentation (its not real indentation the edge just rolls until it is at 90 degrees). All materials which have similar behaviour can then be expected to behave in the same manner. You could include corrosion resistance by cutting wet rope. Some materials are very different yes. For example used carpet is heavy on indentation and wear resistance with rolling as well. However wood induces little wear but a lot of rolling and heavy impacts so fracture is a real possibility. It is easy to see that two blades could have different relative edge holding ability's on carpet vs chopping wood (M2 at ~66 RC vs 52100 at 60 RC for example).

Which edge holding are you talking about, slicing or push cutting?

You can look at both on rope, since they are dependent on similar characteristics you can extrapolate one from the other. I use slicing mainly because push cutting takes longer to do. There are some complications as I mentioned in the other thread (listed below), but these are not first order effects.

Is slicing edge-holding correlated to push-cutting edge-holding?

Yes since it is dependent on the same characteristics. Again there are some complications as with everything, a the recent thread which deals with one such example :

http://www.bladeforums.com/forums/showthread.php?s=&threadid=214104

How do you ensure the sharpness state is the same at start?

You measure it.

You could say that you could design tests for all or most parameters. I'll reply that you'll never know all parameters ...

It is a wedge, it isn't that complicated a tool.

... and that if you take the most importants and rate them, you still do not know if a knife that scores great for all tests is a good knife. In this the human factor IS the key, and surely like our tastes are different, appreciations changes from human to human.

Yes there will be varied opinions as different people put different weights on different aspects, which is why you don't just list the conclusion but what it was based on, what were the criteria used and how did the blade perform at those tasks. It would of course be optimal if the reviewer actually avoided such conclusions, promotional or otherwise.

To clarify, there was no argument being made that all you do is a some rope cutting and then you have a 100% all inclusive view of knife performance. However from looking at how a knife cuts into a piece of rope (when you have done it with many other knives and a lot of other cutting on other materials to correlate the performance), you can form an informative opinion on how it will cut in general.

Baliswinger :

No amount of review reading can compare to how you feel about it.

No of course, reading what someone else has written should never replace personal experience, it should just be complementary.


-Cliff
 
OK, rereading all of this,

It seems that some are talking about doing some \"fundamental\" research on knives, edge profile, blade profile, blade outline, steel, grinds (...), and some others are talking about testing \"a\" knife.

What is the original intention ?

Cliff:

Though I agree that edge-holding as being part of the parameters for a \"high-end knife\", I still think it must fit in an \"acceptable\" range rather than as an absolute value.

May be we should start defining what \"acceptable\" ranges are for common parameters and different uses, but this seems to me also as a very personal scale, the same way that \"high end knife\" is very nebulous to me, as it certainly does not mean the same in my mouth than in yours.

The rope cutting is OK, though correlation to me is not an on off switch, but something that goes from hightly correlated to not correlated. What is the point of getting factual data if it is to empirically decide of corellations afterwards ?

Actually I have to correct myself, as this automated process would test \"conservation of cutting ability\" in function of the amount of material cut, and not \"edge holding\", which is just one parameter of this conservation.

A side question: how do you easily measure sharpness ? (please do not answer the classical hair popping, skin cutting or nail cutting, because if it is done the empiric way, I do not see the point of the automated testing behind).


OK, I stop commenting now, please do it, publish the data, and it will be great.

Cheers,
 
Back
Top