New testing session.

nozh2002 said:
Results are different, so somehow I adopted to testing and start doing it better - one way or another - this need to be analyzed. Is it way I test sharpness, or the way I cut rope - I do not know. But this adoptation need to be somehow accounted. So I can not directly compare old and new results. I guess it should not be much difference in results of the tests done at same time, but defenetly I can not compare recent test with month and a half old.
Click to expand...
Not surprising, this is what I found when I started doing systematic testing ... not only my sharpening skills, but even just my skill using a knife for these kinds of controlled tasks improved significantly.

For your own reference, of course having a few blades that you always include from test-to-test over time let's you judge relative performance. But how to compile all your data over time, with your skills improving and other, not so obvious changes maybe getting thrown in, I'm not really sure.

Again, I really admire all the work you've done, Vassili. Hope you test the Sandvik again ... with a polished edge and microbevel of about 35 degrees inclusive I think you'll see a big improvement.
 
Yes, this is not surprising. The human factor variable has been discussed here before, especially by He Who Will Not be Named. I believe value exists in human knife wielding tests, but somehow the variable has to be teased out of the results to get meaningful information.

I've been thinking about how to set up my own testing apparatus and because of Vassili's recent findings I am leaning more and more toward some kind of motorized jig for holding the cutting material so that it can be brought to bear against a stationary knife edge, one locked in place. I'll have to give this some more thought.

Good job Vassili. This is some excellent information.
 
nozh2002 said:
I just finished retest:

Yuna Hard II ZDP-189 (10 Oct 2008, 1 Dec 2008)

CUT
000 010 010
001 015 015
003 025 020
006 040 020
010 040 030
015 040 025
020 055 030
030 055 030
040 055 040
060 055 050
080 050 045
100 055 050
150 065 045
200 080 050

Results are different, so somehow I adopted to testing and start doing it better - one way or another - this need to be analyzed. Is it way I test sharpness, or the way I cut rope - I do not know. But this adoptation need to be somehow accounted. So I can not directly compare old and new results. I guess it should not be much difference in results of the tests done at same time, but defenetly I can not compare recent test with month and a half old.

I think I need to test Yuna Hard II few more times and then analyze what I will get. Or may be I should retest Sandvic and then Yuna again - having them be retested several times and see what results will look like.

Thanks, Vassili.
Click to expand...

It's a lot of work doing tests like you are doing. And for that, a :thumbup: But I do wonder whether the results aren't almost completely random. The differences between the old test and this new one are enormous. I wonder if there would again be great differences if you did a third test. Variations like this do raise serious questions about whether there's any real validity in this type of testing where a large human factor exists.
 
Elen - That's the problem. We don't know how big a factor the human equation might be until tons and tons of data is produced. Very work intensive. Vassili is to be applauded for his efforts because if nothing else we are seeing just how difficult it really is to produce test results that translate into our respective real world experiences. HWWNBN produced all kinds of test results, but in the end their value is extremely limited because he relies entirely upon human trials and his comparisons ran to only one knife with one edge geometry, and then another test with something entirely different. He did do some relatively close comparisons, but overall this is not easy stuff. If it was, then some manufacturer would have developed a very concise and repeatable test process that anyone could duplicate to show their knife was far and away the best. Nothing out there like that.

Well, there is INFI. :)
 
Elen said:
It's a lot of work doing tests like you are doing. And for that, a :thumbup: But I do wonder whether the results aren't almost completely random. The differences between the old test and this new one are enormous. I wonder if there would again be great differences if you did a third test. Variations like this do raise serious questions about whether there's any real validity in this type of testing where a large human factor exists.
Click to expand...
This is why IMO it's essential to have at least a couple "reference blades" that get used in every round of testing. At least you then have some meaningful point(s) of comparison. But yeah, the actual numbers being generated have no absolute value.
 
Hi Vassili.

I wonder how much the edge angle is influencing the results. I have found that many steels do not hold a 30 degree edge very well. Is there any chance that you would be able to retest a few of the knives with a 40 degree edge angle from a Sharpmaker?


I know this is a lot of work!


Either way, thank you for your efforts.




Frank
 
frank k said:
Hi Vassili.

I wonder how much the edge angle is influencing the results. I have found that many steels do not hold a 30 degree edge very well. Is there any chance that you would be able to retest a few of the knives with a 40 degree edge angle from a Sharpmaker?

I know this is a lot of work!

Either way, thank you for your efforts.

Frank
Click to expand...

There are many factors which may influence performance - if you try everything test matrix became huge. After almost 4 years - I started this about Early 2005 and announce it here May 2005
http://www.bladeforums.com/forums/showthread.php?t=346429.
I still hope more people will join this effort.

Right now I think I need to work on reliability of testing. Right now major consern for me - adaptation - is it has a limit and at this limit does different steel have different results?

If it reach some level and does not advance any more - then I will need to retest early knives until I reach same results - which mean it is point when adaptation stopped. I hope that there is limit for adaptation and so we may have some reliable edge retention results.

I think after I retest first knife, I need to retest one in the middle and see how results will differs and do this again... - do some binary search to find when adaptation stooped (if it did stop).

But even with adoptation stopped - I expect some difference in results and so nee to get median between them as well.

Now I need to give my fingers and wrist little rest. At least today I am not doing anything. Hopefully I will resist for few days...

Thanks, Vassili.
 
Anyone think work hardening could be effecting the results? Or is 200 cuts of manila rope in 24 hours too light of a load to make any noticeable difference?
 
theonew - I'm not sure exactly what "work hardening" means. Does it occur because the material is flexed or manipulated and this creates friction which creates heat? Or does the flexing or manipuation itself realign atoms/molecules, and heat is not an important factor. Either way, I could see how a knife edge (or the micro serrations) could move around quite a bit at the microscopic level as it saws through the rope. This could work harden it to some degree causing the serrations to chip away and dull the edge. Someone else here might have more information about work hardening.

However, even if this is the reason for the dulling effect it would apply across all the blades since the methodology is the same. Perhaps some steel alloys are more prone to work hardening than others and this explains why some edges dull quicker. I am not sure.
 
Work hardening is caused by plastic deformation. Deformation causes dislocations in the matrix which increase hardness & strength, and decrease ductility. Related to fatigue. A steels work hardenability is related to its shear modulus. And small grained steel will work harden faster than one with large grain size.

If you assume these edges are in the 1 micron range, there will be deformation at the edge from cutting, and the steel work hardening through deformation will have a role to play in the eventual wear of the edge. I don't think this micro look at it is the conventional way work hardening is looked at, though...

But there are anomalies in these results if you use the steels work hardenability to rank the steels, just like there are striking anomalies in these results if you look at hardness or wear resistance, which are generally agreed as the two properties that have most correlation to edge retention.

The tests that define sharpness by how much force is required to cut the manilla rope have given much different results - and those results are more predictable when viewed along with hardness and wear resistance. There are undoubtedly reasons for why this is, and I'm pretty sure they could be hard to agree on here! :D
 
Thank you Broos. So it is the deformation and not friction heat that causes work hardening. And yes, I agree, nozh2002's results vary in some places from what might be predicted based on hardenability, hardness and wear resistance theories. Whether that is because of variabilities in the testing methodology or holes in the theories, or both I can't say, but I follow this thread with interest non-the-less.
 
This topic is fascinating to me, and I can’t thank Vassili enough for his time and thought put into his work.

As an engineer I have always considered what testing methodology could be developed to give as meaningful results in this area.

Vassili’s is as well thought out as any I have seen; though the lack of repeatable results always puts any conclusions into question.

One thing I can add is that through my own real world knife usage, cutting a variety of materials on the job site and when hunting/camping, Bob Dozier’s D2 does seem to offer better edge retention then any other steel I have used.

Keep up the good work Vassili; I look forward to seeing what you come up with down the line.

And for the rest of us, Vassili has set the bar pretty high; we should all strive for such a scientific approach to our own testing.





"If you're not living on the edge, …you're taking up too much space."

Big Mike
 
Big Mike said:
This topic is fascinating to me, and I can’t thank Vassili enough for his time and thought put into his work.

As an engineer I have always considered what testing methodology could be developed to give as meaningful results in this area.

Vassili’s is as well thought out as any I have seen; though the lack of repeatable results always puts any conclusions into question.

One thing I can add is that through my own real world knife usage, cutting a variety of materials on the job site and when hunting/camping, Bob Dozier’s D2 does seem to offer better edge retention then any other steel I have used.

Keep up the good work Vassili; I look forward to seeing what you come up with down the line.

And for the rest of us, Vassili has set the bar pretty high; we should all strive for such a scientific approach to our own testing.





"If you're not living on the edge, …you're taking up too much space."

Big Mike
Click to expand...

Thanks Mike for your words of encouragement. I really need them. Now I made some pause to let my arm rest and recover before I damage it too much and will have to wait another year.

Right now tightening thread on the base (the right gadget on this picture) is what cause me trouble.

Random-420.jpg


14 tests 21 thread cuts each (294) for 26 knives - I did it 7644 times and this is too much for my palm. I need some easier way to do this, but can not came up with any yet.

Thanks, Vassili.
 
what about hanging a weight from the thread to provide tension. You could put some sort of spring clip on the weight, so you don't have to tie knots. Or clip it to the scale and preload it.
 
Last edited:
I am thinking about something like this.

springsys.jpg


Any idea from what I can made it?
Where to take this parts from - I prefer first one, but where I can get metal good enough for this cause?

2 hardheart - problem with weight is that it will drop any time I cut thread. I did something like this with weight and placing knife on scale and moving thread instead with weight on one side... Some construction I build around March this year, but it did not work out.

Thanks, Vassili.
 
Ingenious,

…but it looks like a lot of work for the number of times you have to re-set it.





"If you're not living on the edge, …you're taking up too much space."

Big Mike
 
You must log in or register to reply here.
Back
Top