The Best Steel - Cut Tests and Opinions

Status
Not open for further replies.
There have been some pretty strong opinions voiced above...sort of surprised me. The OP tried to give us some sort of compilation of data that he had found. I found it interesting; his efforts are appreciated, but I generally fall back to Ankerson. I do thank the OP and will stop short of pounding his efforts into the dust...regardless of how "accurate" several of you other folks think you are... "Beware of karma...it'll certainly kick you in the butt..."
 
knarfeng said:
I completely disagree. Erroneous data is worse than no data.
Click to expand...

And an invalid controlled experiment is worse than erroneous data. So often these "scientific proofs" wouldn't pass muster in a 6th grade science fair.

I mean nothing wrong with anecdotal observations/conclusions, but let's not pretend they are something else than what they are.
 
Sonnydaze said:
AGREED!
Click to expand...

If I had a nickel for each time I saw someone "scientifically prove" stuff about steel. It is super, super difficult to come up with a controlled experiment. I mean, if the experiment doesn't start with Spyderco Mules, then it is pretty much doomed from the outset.
 
I like to get a knife and use it to make my own assessment. Someone else’s data is simply that and when dealing with production knives there’s way too many variables.

What gets me is the strife for the perfect knife and steel to not use it and resell it when the CC charge hits. At this point, I’d assume more than 3/4 of ‘high end’ knives will never see use and never be judged for how well they work.
 
"Spyderco’s Mule Team Project is one of the most unique programs in the commercial knife industry. By definition, a “mule” is a sample knife used for in-house performance testing. Traditionally, they are made to the exact same pattern and specifications, but feature a different blade steel or heat treatment protocol. This keeps all the performance characteristics of the knife identical except for a single variable, allowing an excellent basis for objective, scientific performance-based testing."

Darn tootin'.
 
knarfeng said:
I completely disagree. Erroneous data is worse than no data.
Click to expand...

Disagree, especially if we understand how much error there is and why there is error. An easy example is systematic error, let's say we calibrated the scale wrong and lost 1lb of cutting pressure so we stopped the test early because it read 1 lb heavy. We could then gather information from the erroneous result and test a hypothesis. Further, if we only knew the scale was off by a 1lb we could then set error bars of +/- 1lb variation. The data becomes increasingly unreliable but trends can still be made but any data points that are close to each other give us inconclusive information.

From the cut tests from the Cedric and Ada, we could conclude that the CPM M4 cuts really well and 420HC does not, in a relative setting. As the differences become closer the error range then starts to blur the differences to where the actual numbers, assuming infinite iterations of the test were performed, not the measured ones, would be inconclusive. We can still answer some questions there's just lower resolution and some questions that we may have wanted answered that can't be answer with a proper level of certainty.

It's really a question of error vs no-error, it's a question of how much error as it's quite rare that a scientific study is error free, thus the propagation of error practice in scientific circles, something I'm admittedly bad at but at least understand the concept.

The information posted by the OP from Cedric and Ada certainly has errors and it's far from the most scientific study but it has a hypothesis that is tested and results that can create a trend of information. It wouldn't pass a peer review for a scientific journal but certainly enough for comments on a public forum and enough for someone to see semi-empirical evidence showing super steels holding an edge longer.

Edit: this isn't necessarily directed at you as it's explaining my thought process on one side of the debate over the validity of the info the OP shared.
 
whp said:
OP, thanks for this thread. The identity of the knives used tomtest various steels provides addtional useful information.
Cedric and Ada s information may not be scienfific, but it is not useless. These comparisons may have some correlation with real world use.
Click to expand...

Yes, the tests are somewhat accurate. All that you need to do is to look at the list and they are ordered according to number of cuts of 1 inch thick sisal...based on:
- edge geometry
- heat treatment of the steel
- error

Now mind you, I said that within plus or minus 10 to 15 cuts, that is my rough estimate on Pete's error. He is not a robot, so, I'll give him that margin of error.

If you look at any reasonable steel chart arranged hierarchically for edge retention, Pete's list matches it quite well!
 
me2 said:
Or, he gets what he expects based on all those other charts.
Click to expand...

Certainly a possibility, especially given that the number of cuts is determined by "when the knife can no longer cut printer paper."

That's certainly not in any way scientific/objective.
 
Some of you guys seem really troubled that anyone would use these rope tests to make a conclusion about a knife or type of steel. First of all, it’s up to the video viewer to utilize the presented information in a way that they personally see fit. If they use the information to come to a conclusion and it turns out that the conclusion was wrong, well, oh well. If I think one type of steel is better than the other and I’m wrong, I’m not going to lose any sleep about it. You guys make it sound like the results of these test will have life and death consequences instead of some guy that is maybe using it to decide which knife he might want to purchase.
 
Snipe 1 said:
First of all, it’s up to the video viewer to utilize the presented information in a way that they personally see fit.
Click to expand...

Precisely. And I am using it to point out the weaknesses in his method.

If you want to use it otherwise, that, by your own claims, is your right.
 
knarfeng said:
The edge retention ratings are useless without Rockwell readings.
Two blades made of the same alloy can have vastly different edge retention results if the blades are hardened to different hardnesses. Without Rockwell data one cannot make valid comparisons.

I've seen his test method. He cuts on a wood surface. The amount of edge damage from cutting into the wood is a variable which can impact results. A better method is to cut so that only the rope is being cut.

These errors lead to a lot of his results not making much sense. For instance, he lists 4116 in the same category as D2. 4116 is essentially 420HC. Putting these two alloys in the same performance category is laughable.

It's always amusing to see testing and conclusions from people with no technical background.
Click to expand...

This.

And even at the same hardness identical steel still can behave very diffidently due to the internal structure... the size and the place of steel grain, carbide, characteristic of martensite formed etc. are all have so much influence.
 
"Just curious: Aren't you a bit worried about making such ridiculous pronouncements for something that was published less than 20 years ago in a major knife magazine publication? One other poster here did remember it, but only vaguely. Could you come up with reasons I would make this up?

Haven't you at least considered what will happen when it turns up?"

I'm not the slightest bit worried for a couple of reasons Gaston. First of all you have had almost 5 years to come up with it and can't. I had pretty much all the knife mags back then and finally threw them away just 4 or 5 years ago. If that was in there I would have seen it.

When I'm wrong about stuff I learn from it and don't take it personal and decide to deny, deny, deny and hold a grudge forever. I thank the people who gave me more knowledge rather than think of them as an enemy for pointing out where I was wrong.

I pointed out the article you used as the basis for your memories. I pointed out it was 440V and not 440C that led the way in wear resistance testing. An honest mistake anyone can make but most don't have a pathological inability to admit being wrong.

If you could find an article from that time frame that said essentially what you claim it said I would look at the article to see what I could learn. I would also check the changes and progress made to our knowledge base over the last couple decades and see what still applies. If I had questions about it my head wouldn't spin and fire out pea soup. Instead I would see if I could replicate the things I questioned, see if there are others who have and look at their results. All in all I like to think I would be practical about it and my thought process would follow some form of logic using accepted and simple problem solving techniques. Then I would sit down to a cup of coffee and watch a movie. :)

Joe
 
Gaston444 said:
Cold Steel claim 4116 out performs ALL 440s:

"4116 is a fine grained, stainless steel made by ThyssenKrupp in Germany and is used for hygienic applications (medical devices and the pharmaceutical industry) and food processing which make it a superb material for kitchen cutlery. The balance of carbon and chromium content give it a high degree of corrosion resistance and also impressive physical characteristics of strength and edge holding. Edge retention in actual cutting tests exceeded blades made of the 420 and 440 series of stainless steels. Other alloying elements contribute to grain refinement which increase blade strength and edge toughness and also allow for a finer, sharper edge."

Also:

"Alberta Ed
5,574
Jun 29, 1999
There are so many factors affecting 'edge retention' that claims of one steel over another are kind of meaningless. I believe that 420HC out-performed 440C in Buck's CATRA tests, and that is one reason why they chose that steel. I have a bunch of CS knives, including some in mystery '400 series' stainless steel, and they all cut."

As far as chopping wood with a 10" blade is concerned, 420J2 is definitely way, way better than Lile's D-2 in my experience: Very comparable in initial fine edge holding to D-2 (both excellent), yet 420J will never chip (and this on a $100 knife vs $1700 in D-2, both nearly identical in shape). For some reason cheap 420J does not have any propension to micro-roll under impact either... I would agree that, for me, 420J does outperform 440C in fine edge holding (no micro-rolling) while chopping, 440 only coming close with a micro-bevel, vs 420J with none: Amazing.

Only -30 C temperatures seemed to be 420's Achilles heel....

As I pointed out many times, the Knives Illustrated magazine test of 1999 had 440C way ahead of everything else, including early CPMs (including 3V) and INFI: Only D-2 came anywhere close in the rope cutting, and it was far behind... The 420J I have definitely behaves like it is better than 440C, and Cold Steel's claim for 4116 seems to be in line for that: Could be all about the heat-treat...

All of that with 420 being far easier to sharpen... If 4116 is anything like 420J or HC, supposedly "amazing" claims about it are not entirely impossible. Theory only goes so far.

Gaston
Click to expand...

The very first sentence is inaccurate. In the quote from Cold Steel, they claim it outperformed (some) blades in the 420 and 440 class. The word in parenthesis was added by me to show the error. Also Everybody already knows D2 doesn’t make the best chopper steel. 420 does not come close to the edge retention of a large(13”) chopper in infi. And infi is practically stainless, though not by definition.
 
Status
Not open for further replies.
Back
Top