Performance testing?

[sargey]

hi again cliff, how're you doing?

Yes, and it will be very precise. Meaning the results will show little variance. This is however only half of the equation. The accuracy, or how close the measured results are to the quantity they are supposed to be estimating, depends on completely different factors.

If the goal is to decide which steel works best in machine cutters, that is an excellent test. However if it is to predict which steel works best in a human hand, it is a pretty lousy test because it ignores critical aspects of a human hand.

we disagree again it seems, the goal was to come up with a means of comparing the performance of different knives to each other, developing a test that could be repeated by makers in their facilities several hundred miles apart. by measuring the pressure required to repeatedly cut through the substrate you are comparing the accumulation of various aspects/compromises of blade design ie steel, heat treatment, geometry as i have already explained. that much at least, cannot be denied.

as for the ergonomics, and factors that affect the cutting ability, your comments about the test ignoring critical aspects of the human hand don't make any sense. the blade cuts, with either a chopping or slicing movement. obviously a human will not make exactly the same cut every time, we don't want to measure the human's performance, we are trying to measure the blades performance. there are many other factors to take into account, the handle, the grip, the material, the angle between blade and handle and so on. this is why it was suggested that we have two different classes of test. repeatable, unsubjective tests in the lab, and real world, real people outside.

i understand your arguments pertaining to development/tweaking of the breaking system. it isn't a viable analogy, we're not using this as a development tool. you could use it as a limited development tool however, as mentioned earlier. it could be used to compare different geometries manufactured from the same steel and heat treat, you could compare different steels with the same geometry and so on. if you were qualified or knowledgeable in the design of experiments you could possibly compare more than one variable simultainiously, but the issues involved are too complex for us to bother with at the moment.

To clarify, my main point isn't against controlled testing. I think it can be very valuable, and materials properties are the place to start. The next point would be getting makers and manufacturers to be more public about the blade specifics, balance, edge geometry and so on, and then get some rough standards for performance in place. Once these start to get accepted, then they can be refined.

here i agree wholeheartedly, if we could develop some meaningful standards, our golden/platinum age of cutlery could only get better. the major winner has got to be the poor budget challenged consumer!

cheers, and.

 

[Cobalt]

Wow, a long thread and I'm not in it. So here goes.

First, being a mechanical engineer, I agree that some sort of standardized testing would be beneficial to comparing knives. The problem is that there is no such thing. Many knife co.'s make their own tests, like CS, Falkniven, and whomever else. Additionaly, the testing would have to be done by non-biased testing lab like Underwriters Laboratories (UL) or Don King promotions.

Who is going to decide what tests are legit?

I remember Jerry using Cold Steel's testing to compare the BM to the TM. Of course the BM blew the TM out of the water in those tests and the results were confirmed by other individuals including Clif.

So lets decide what testing should be included or done and specifically how?

I will add the first test. Knife must be able to cut all of Don Kings hair without dulling.

 

[Jerry Busse]

"ALL" of Don King's hair????


Okay, I'm out! That test is way too frightening!!!!!!!!


Holy cow man! It's a knife. . . not a weed whacker!!!!

 

[Matteo Escobar]

The bottom line is- if you break it, Jerry will replace it. Go ahead and do whatever the hell you want with it. As stated, of course a thinner knife will slice better than a thicker one. And of course a thicker knife will pry better than a thinner one. It's not complicated. It all comes down to steel, heat treat, and ergos.

 

[Cliff Stamp]

sargey :

the goal was to come up with a means of comparing the performance of different knives to each other, developing a test that could be repeated by makers in their facilities several hundred miles apart.

This is only half the picture. The tests must also be useful as well. You need accuracy as well as precision.

by measuring the pressure required to repeatedly cut through the substrate you are comparing the accumulation of various aspects/compromises of blade design ie steel, heat treatment, geometry as i have already explained. that much at least, cannot be denied.

The geometry is indeed tested by how much force is applied, and some properties of the steel influence the edge lifetime. However some properties of the steel are ignored, and these ones are in fact critical to the blades performance when used by a person, just as in the example of the robot break testing in the above.

In fact the testing you have described has already been done by CATRA. It also has the exact same problem I described. When Buck tested their Ionfusion blades on CATRA machines, the results were insane, like 1000 times edge life. In actual use, the advantage was not even close to this, the performance was off by orders of magnitude.

Just consider what this means. It would be like asking how much a knife is, getting an answer of say $100 give or take, and then getting your VISA charged for 50 000$. More than just a little bit off.

How come the CATRA use doesn't correlate well to people, because people are not CATRA machines and use blades differently. You can't test for different properties and have a meaningful test regardless of how precise you are.

Buck realized that these results were not meaningful and stopped using the machine and instead had people do some cutting to determine edge life as the machine was too artificial as it ignored factors critical to edge retention when the knives were being used by people.

we don't want to measure the human's performance, we are trying to measure the blades performance.

And that is where the test fails, because humans are using it. You can't ignore factors critical to performance and expect to get an accurate representation of an effect.

i understand your arguments pertaining to development/tweaking of the breaking system. it isn't a viable analogy, we're not using this as a development tool.

Fine change the description to using the robots to test breaking systems as an aftermarket rating system (consumer reprorts). It still fails horribly.

if we could develop some meaningful standards, our golden/platinum age of cutlery could only get better.

Start asking, in fact demanding, for them from the makers / manufacturers. Get some pressure on them to do the testing, it isn't sensible to expect the consumer to do it. Especially considering how restrictive some warrenty policies are on the blades.

-Cliff

 

[sargey]

we could be here chasing each others tails for quite a while i reckon.

can you tell me what the numbers next to the volume knob on my stereo refer to? why don't the numbers on a vickers or rockwell hardness scale match? why do we have both faranheit and celcius temperature scales? frankly, who cares? it's a simple frame of reference. a benchmark. you cannot predict in any tests, the skill of the end user. it could be a kendo master with jedi like abilities who never misses a stroke, or captain clumsy with a rock magnet in each wrist, the guy who cannot cut a sapling without digging up a few chunks of granite on the follow through. it is obvious that some folks will be harder on their knives than others, it is also impossible to predict. if all the knives are measured or graded, or given a star rating in the same manner, as per an earlier post of mine. the consumer has a simple idea of how this knife performs relative to other knives. it really is that simple. there is no point in attempting to make more of it than that.

now, there has been no suggestion that this pressure/cutting test is the be all and end all of knife testing. it is just one tool, to be used in conjunction with others, to give the end user a verifiable idea of what they are buying. it would be folly to rely on this one test on it's own as the measure of a survival knife. just as it would be madness to judge a knife on it's ability to withstand a lateral stress experiment, to the exclusion of all others.

i'm sorry cliff, but i fail to understand your attempts to read so much more into this side of the argument than there is. your robot analogies make no sense. i am not suggesting comparing a breaking system to a robot, i am not suggesting asking a robot's opinon in an aftermarket survey. i am suggesting that we compare one knife to another knife in a repeatable, verifiable manner. wether those knives are rated "marks out of ten" or on a percentile basis, or 1 to 5 stars, is largely irrelevent. you can set a scale anywhere you like. wether the scale is vickers or rockwell doesn't matter. as long as the scale is fixed. your insistance on having measurements applicable to the user is like measuring the distance on a map in steps! how do you intend to predict a hikers length of stride???

it's late here, and time for bed. cheers, and.

 

[Ebbtide]

If I were building brakes....
I'd test them in an even careful manner...
Then in panic stops.
Then long downhill foot on the pedal runs.
Followed by panic stops.
Etc, Etc, Etc.....


One test does not do it all.
Sargey's device is one.
Bagwell's "Hit it on the horn of an anvil" is another.
Rope/tomato/pop can cutting are some more.

Then from all of the controlled tests together we can form our own opinions.

And we all know about opinions...

 

[UffDa]

See, here we are on page 4 of this thread and, as I said early in this thread, there are some who will never be satisfied. NEVER!!!
It just goes on and on and on and on........

The opinions expressed here are interesting and informative, but accomplish nothing. Well, it is better then watching reruns on Animal Planet.

 

[Cliff Stamp]

sargey :

you cannot predict in any tests, the skill of the end user.

Which is why you would ideally measure the performance over a range of skill and physical strength levels to show the influence, and provide a few simple ways for a reader to estimate their abilities and thus the performance of the knife in their hand.

At a very basic level though you would not even need to do the second part. For example for edge retention on fillet knives you go down to a wharf and pass out some knives to a lot of fisherman and gather some results. You then make a report like :

Knoara brand -> 10 to 30 fish
Yorhee brand -> 15 to 50 fish
Billaw brand -> 5 to 15 fish

The guys using the fillet knives ranged in experience from complete novices to experienced fisherman with more than twenty years of handling cod.

A reader can thus look at the above and just from his own experience and what he wants as to what is suitable. A novice for example might choose the Yorhee because he needs all the help he can get. A very experienced fisherman might just choose the Bilaw becuase he knows he will be 15 or so more and then a quick steel and he is off again.

Of course there would be lots more to consider, handle ergonomics, security when greased, flexibility, ease of penetration of the tip, corrosion resistance, resistance to chipping, etc. .

if all the knives are measured or graded, or given a star rating in the same manner, as per an earlier post of mine. the consumer has a simple idea of how this knife performs relative to other knives.

When used by a machine. Unless the reader is also a machine they can't extrapolate their results. This has already been tried and fails horribly as I described in the above.

Do you want to use it to measure the effects of geometry, great it does that, and for some things it is ideal, like penetration tests and so on.

However if you want to use it to measure the effect of edge retention (or durablity) it is horrible because it ignores factors which are critical to edge retention when the knife is used by people.

This is the exact same problem I described with the breaks. You can't use a robot to judge the relative performance of breaks unless you want robots to be driving cars.

your insistance on having measurements applicable to the user is like measuring the distance on a map in steps!

This is simply a change of units, you would just scale to whatever you were comfortable in. The test you describe ignores critical factors of performance and can't be scaled at all.

-Cliff

 

[sargey]

the fillet knife test works very well, because there is a specific application. it is relatively easy to design a tool for a very specific application, and it is relatively easy to verify that tool's performance at that specific application. it can also work well, as it seems that you propose having a substantial sample size.

now say we have a knife, more to the point, a whole range of utility/survival/hunting/camping/combat knives, with a whole range of very different end users and end uses. you cannot feasably design tests that will take all of this into account, you can't do in a repeatable manner, or in a way that is economically viable for an industry to support it. if you can, please explain how.

i have proposed a rig/test of sorts that will produce a consistant result when products are compared in a consistant way. it seems that someone else has already done the same, what can i say, great minds think alike you have suggested that this tool will be out by several orders of magnitude. this consistant result is out by several orders of magnitude. you have already scaled it up, and yet you go on to say that this consistant result cannot be scaled down. then you suggested that your subjective variable measurements can be scaled. you are not only contradicting yourself, you seem to be suggesting that we define the answer to "how long is a piece of string?" by measuring it with a rubber band!

i think that there is a major flaw in this debate as a whole, we have the catra equipment, and we have the test rig in my head. if we had a working model, we could develop it to do anything, this debating the qualities of a hypothetical machine is proving to be largely fruitless. i am a working engineer, i have to produce results that are both viable and achievable, i have to be a pragmatist. was it macarthur who said "better to have a good plan now than a perfect plan later."?

i understand your concerns around the issue of making claims in the lab that cannot be clicked and dragged into the real world. you suggest that the cutting pressure test is of no practical use. so, how about hypothetical ratings A to F across three categories, penetration, lateral stress and cutting pressure. A is world leader class, an F is negligable performance. i would like to have a reliable repeatable test for chopping as a fourth class. but off hand i can't think of one.

suppose that a battle mistress rated B.A.D and that a 10" fallkniven rated B.B.C. a mora on the other hand might rate A.E.B. a scalpel might rate A.F.A. doesn't this immediately say loads more about the tool in question than that other famous accepted benchmark, the rockwell test? what good is a measure of hardness to a consumer? it doesn't tell you anything about the knife on it's own! it is of no practical use! at least not without a load of extra data.

i had sort of hoped that this thread might have produced a set of tests or parameters that some folks could have agreed on, then we could have moved on to refining them. it does seem to be getting bogged down. and as uffda said, not achieving much. i'm off on holiday this weekend, (friday) and i'll have no i'net access while i'm away. but i'll check in when i get back. see ya then.

cheers, and.

 

[Ebbtide]

I believe General Patton said:
"A good plan, violently executed now, is better than a perfect plan next week."

 

[Cobalt]

Originally posted by Jerry Busse
"ALL" of Don King's hair????


Okay, I'm out! That test is way too frightening!!!!!!!!


Holy cow man! It's a knife. . . not a weed whacker!!!!

oh, yah, sissy boy, you afraid of a little DK Hair. Well, put up or shut up. By the way the second test involves Mike Tyson biting the Tang off.