Steel Edge Retention testing

[HoB]

that being said then a thinner knife should last longer for the same steel used.

I think the problem here is that you do not distinguish properly between edge thickness and edge angle. When we talk about edge thickness, we mean actually the thickness behind the edge. Nozh is not carrying out the tests to a stage where wear is so significant that the thickness behind the edge will affect the test. I might affect the cutting of the rope, as a thin edge might go easier through the rope (but then again, rope is very plyable, so the effect might be fairly small), but not the sharpness test afterwards, because at the very edge, both thick and thin edge look the same as the have the same edge angle and on thread cutting only the very edge comes in contact with the thread. (Which is btw. what Nozh has been saying all along ). No offense, but at the moment, I think, few people have as much experience, cutting rope than Nozh. If you go to his page and count out all his tests, it amounts to, what, close to 10,000 cuts through rope?

I agree with Larrin, I don't think there will be much destinction between thin and thick edge. It think with those steels that have an appreciable carbide fraction edge angle is really the key factor. I will dare to postulate, that these steels will perform very well for edge angles that are normally found on folding knives (10-12 deg per side minimum). I would expect that their performance will probably come crashing down at really low angles though. And not because of the thread testing afterwards, but because I suspect that you will experience carbide tearout during the rope cutting at very low angles.

 

[NuclearBossHog]

I think the problem here is that you do not distinguish properly between edge thickness and edge angle. When we talk about edge thickness, we mean actually the thickness behind the edge. Nozh is not carrying out the tests to a stage where wear is so significant that the thickness behind the edge will affect the test. I might affect the cutting of the rope, as a thin edge might go easier through the rope (but then again, rope is very plyable, so the effect might be fairly small), but not the sharpness test afterwards, because at the very edge, both thick and thin edge look the same as the have the same edge angle and on thread cutting only the very edge comes in contact with the thread. (Which is btw. what Nozh has been saying all along ). No offense, but at the moment, I think, few people have as much experience, cutting rope than Nozh. If you go to his page and count out all his tests, it amounts to, what, close to 10,000 cuts through rope?

I agree with Larrin, I don't think there will be much destinction between thin and thick edge. It think with those steels that have an appreciable carbide fraction edge angle is really the key factor. I will dare to postulate, that these steels will perform very well for edge angles that are normally found on folding knives (10-12 deg per side minimum). I would expect that their performance will probably come crashing down at really low angles though. And not because of the thread testing afterwards, but because I suspect that you will experience carbide tearout during the rope cutting at very low angles.

Ok, so your saying that at very low angles the performance will not be good. What angles are you talking about? something like 18 or more? Dont these angles translate to a thicker edge? Isn't it the same thing?

In any case, I think that this being the case, then testing at different "angles" is important for each steel.

Here is example of what I mean. I have two Beckers, a BK9 and Brute. Both are 9 inch knives and both or totally different. My BK9 has very thin edge, I would almost bet it is 14 deg. My Brute has much thicker edge probably close to 18 or more. My BK9 will outlast my brute on general cutting chores. I have used both to cut cardboard, rope and tons of cloth. But when I have used both on chopping hard materials, the edge on the brute is much more durable and I have not had anywhere near tghe chipping issues that I have had with the BK9 and both are the same steel, same hardness, same heattreat and same maker. The thicker brute edge doesn't damage as easily but wont last as long. why? The only difference is edge angle, ie. edge thickness. I am pretty sure that if I did just a straight cutting test the bk9 would beat the brute hands down. I have already done it without taking notes and have seen it.

 

[Any Cal.]

Degrees are a measure of edge angle. The measure of an edge's thickness would be stated in thousandths or mm. Multiple edge angles can give a similar edge thickness, depending on the stock that the blade is made from. A thin knife w/ a keen edge angle, but scandi ground, will be the full stock thickness at the edge. A knife w/ a secondary bevel can have a more obtuse edge angle, yet still have a thinner edge, due to where the top of the sharpened portion meets the grind.

 

[annr]

Hey NuclearBossHog!
I tried sending you an email and it bounced back. If you want to shoot me an email via BFs I can try to resend.
annr

 

[HoB]

Edge thickness and edge angle are not identical, this is why we have this discussion. As I have stated before, the term "edge thickness" actually refers to the thickness BEHIND the edge (see sketch). You can have at the same edge angle a small width behind the edge bevel or a large width behind the edge bevel. You can further play with this by introducing multiple (micro bevels). Since in thread cutting you don't have penetration, thickness is not a quantity to be considered for the sharpness test, only edge angle. This is what Vassili has tried to say all along. The "thickness" of the actual edge is impossible to determine without an scanning electron microscope (SEM), so it is hardly a useful quantity for the average knife user. Since the edge is not flat, one would speak actually of the radius of the edge curvature and that is for a sharp edge in the 0.1 micro (for extremely sharp razor edges, I doubt that more than a few people here have ever honed an edge that fine, me included) up to 10 microns where a shaving edge more or less ends. And roughly 50 microns for a just so working edge (these are ballpark figures only!). The edge radius is not only depending on edge angle but of course also on how well you hone the edge, what grit you use and whether the material supports such a fine structure. But Vassili is eliminating most of these variable by honing consistently to a very high grit level, so the edge radius is mainly dependent on the edge angle. After a few strokes the very high sharpness is gone anyway and so a slight variation in edge radius is irrelevant anyways. The edge angle remain still an important figure as it determines how much material you have in the first few microns right behind the very edge.


Well, I have no idea at what angle you might reach the critical point, which will be of course a characteristic quantity for each steel and be dependent on carbide size, distribution, and volume. If you read my post though, I stated a number. I know from personal experience that a steel like S30V can perform quite well at an edge angle (no microbevel) of 12 deg (per side). I know that several people had problems with S30V at something like 5 deg per side. So somewhere in between seems to be the critical angle for S30V. I don't know what your "18" is referring to, nor do I know whether you are talking about include angle or per side.

 

[roughedges]

OK, eventually, I'll do this test. But thickness of the blade is not important for edge retention testing. Period.

Thanks, Vassili.

In my opinion, it is important if your intent is to compare the steels themselves and not any specific knives made in those steels. Basically, its a simple need to control as many variables as possible in order to isolate the ones that you seek to compare.

I sincerely applaud what you've done, Vassili, but it is mostly a comparison of specific knives and not the steels that they are made from. In my opinion, this does not in any way lessen your research but I think it is something to consider. To me, the ideal for tests like these would be something like the spyderco mule project knife in various different steels. Same geometry, same edge angle/thickness, different steels. This would give you a true comparison of different steels and not the knives employing them.

Regardless, keep doing what you're doing. I enjoy it.

 

[nozh2002]

In my opinion, it is important if your intent is to compare the steels themselves and not any specific knives made in those steels. Basically, its a simple need to control as many variables as possible in order to isolate the ones that you seek to compare.

I sincerely applaud what you've done, Vassili, but it is mostly a comparison of specific knives and not the steels that they are made from. In my opinion, this does not in any way lessen your research but I think it is something to consider. To me, the ideal for tests like these would be something like the spyderco mule project knife in various different steels. Same geometry, same edge angle/thickness, different steels. This would give you a true comparison of different steels and not the knives employing them.

Regardless, keep doing what you're doing. I enjoy it.

Well, you are wrong. To test steel edge retention there is no need to have identical knives from all awailable steel. Reason why this is wrong and why this test show true difference between steel (from given manufacturer) edge retention at 30 degree edge angle (15+15) was explained in great details by me and by many others already many times, so I will not go over this again. Especially after it was done right before your post.

There is no need to have identical knives to test edge retention. No need to have same blade length, no need to have same blade width, no need to have same spine thickness, no need to have same bevels and no need to have same edge thickness for this particular test (while this thickness bigger then thread thickness).

Of course this is not steel test but steel from given manufacturer test at given edge angle on 1/2" manila rope - heat treatment may be different from manufacturer to manufacturer and this is why I say what knife do I use, different media may affect edge differently also, as well as edge angle may affect the way edge wear out. It is very possible that from manufacturer A steel S1 will be better then S2 but from manufacturer B it may be very well steel S2 better then S1 - however I doubt that it is too narrow. I think some may do better then other but not to much - we'll see.

So I can see test matrix where we have same tables for same knife but different edge angle and for same steel but from different manufacturers, or even different HRC... Like 3 dimentional matrix...

But when nobody helping me with real work, nobody joining my effort, I guess you all have to wait until I finish what I have in my mind for next. And my plan is to test different steels first from top manufacturers with same angle.

I really wish someone will do same test, until then - it is all not quite "scientific". Why can not we have some knives open community project like so many projects in Linux world?

Thanks, Vassili.

 

[theonew]

Vassili

I'm so close to jumping in on this. I just wish there was a quicker way to assess sharpness, rather than having to make 20 thread cuts. I also feel a strong need to assess a margin of error along the lines of testing the exact same knife with the exact same bevel against itself or two knifes with same steel but different spine widths or different edge bevels.

 

[sodak]

But when nobody helping me with real work, nobody joining my effort, I guess you all have to wait until I finish what I have in my mind for next. And my plan is to test different steels first from top manufacturers with same angle.

I really wish someone will do same test, until then - it is all not quite "scientific". Why can not we have some knives open community project like so many projects in Linux world?

Thanks, Vassili.

There are some open community sharpness and edge testing experiments going on right now, I currently have the knives that are being passed around. I've been slow finishing the testing, but will get them completely tested in the next couple of weeks, then on to the next person. I will post all of my results and test methodology in detail, if you are interested, please pm me and I'll see about getting you added to the list of people.

 

[nozh2002]

Vassili

I'm so close to jumping in on this. I just wish there was a quicker way to assess sharpness, rather than having to make 20 thread cuts. I also feel a strong need to assess a margin of error along the lines of testing the exact same knife with the exact same bevel against itself or two knifes with same steel but different spine widths or different edge bevels.

I wish to have quicker way to assess sharpness too. For sharpening I do hair whittling but if we want to see dynamic of edge degradation we need more precise way, which shows this in numbers. I do not know may be less then 21 thread cut will work, but there is no big difference between like 17 and 21 in terms of workload. However I am already doing this for some time - I developed pretty thick skin on my fingertips, etc. For me now it is pretty easy to do.

Thanks, Vassili.

 

[nozh2002]

There are some open community sharpness and edge testing experiments going on right now, I currently have the knives that are being passed around. I've been slow finishing the testing, but will get them completely tested in the next couple of weeks, then on to the next person. I will post all of my results and test methodology in detail, if you are interested, please pm me and I'll see about getting you added to the list of people.

Well, I am kind of busy doing this tests and do not think can really join another effort - already too much to do for me alone. Sorry.

Thanks, Vassili.

 

[roughedges]

Well, you are wrong...
But when nobody helping me with real work, nobody joining my effort, I guess you all have to wait until I finish what I have in my mind for next. And my plan is to test different steels first from top manufacturers with same angle.

Fair enough, I suppose. good luck with your testing.

As for the second part of what I've quoted above, many people have offered you a lot of advice. You've chosen to disregard, out of hand, any that you do not immediately agree with. Advice given from several laypeople and one acknowledged expert in the field.


HoB may be spot on, and the edge angle and thickness might not matter much in a test like this, but, as long as you are not controlling those variables, you are testing the knives themselves and not the steel that they are made from. It is as simple as that.

I will say it once more. I am not knocking your tests and I applaud what you've done. Good luck with your continued experimentation.

 

[nozh2002]

many people have offered you a lot of advice. You've chosen to disregard, out of hand, any that you do not immediately agree with. Advice given from several laypeople and one acknowledged expert in the field.

Well, I see this different way. To me it looks like people raise their opinion - like you did and I explained many times and in great details my opinion and thoughts behind it, other members too - HoB and db and AnyCal. In replay I only see same statement repeated and no discussion.

Correct me if I am wrong. To me - I explain my point many times and it were my opponents who choose to disregard my and others willing to discuss it and our arguments.

More then this - you for example did not even see my coomet on angle - and I directly sad I think it is important and I have all blade sharpened under same angle.

I do not accept anybody as a higher being and I listen only to arguments not to title. For me it does not matter do someone known as a expert until he able to communicate his expertise and will to discuss matter based on his expertise with "regular" people. I got enough directions before, I am deciding myself what to do and if you can not explain and discuss your advice to me and I have unanswered questions - sorry. But you alway welcomed to do it yourself - I will do my best to explain how all this works.

Thanks, Vassili.

 

[Don Powell]

Vassili, I applaud you. I applaud your effort and labor, and I applaud your wisdom and insight to continue on this quest in the way that you have. Let the detractors, with their nits to pick, devise a test that will encompass what they think is correct and cut as much rope as you have, then they can come back and try to discourage your effort.
Best wishes, Don

 

[Broos]

I have seen before when test results are posted & comments are solicited, and then feelings get hurt when comments, suggestions, or criticisms are offered. It may be best to accept that when posting tests here that everyone will not agree with all of your methodology or conclusions. There is not enough accepted theory or test standardization to think that everyone will agree on what parameters are important, the nature of the edge degradation, or what conclusions can be made from test results. To my thinking, the test defines the conclusions that can be made, in that I would define my conclusions based on the test method, and that is enough.

Thanks again for sharing the results!

 

[nozh2002]

I am fine with one or other interpretation of the test. I like discussion with arguments, contr-arguments and contr-contr-arguments. I am open minded and listen to arguments others have and accept them or came with contr-arguments or explain my position more if I see it may be not well understood.

This way all together we can finally discover the truth.

But if there is no discussion, there is no arguments and contr-arguments, if it is just statement repeated, no will to listen to opponent no will to discuss opponent argument - nothing I can really do. I respect position of other people, but I also respect my position to give it up just because some well known expert disregard it without any explanation and ignoring my effort to explain.

So let respect each other position and let me continue with my exercise.

Thanks, Vassili.

 

[LFree]

There is no need to have identical knives to test edge retention. No need to have same blade length, no need to have same blade width, no need to have same spine thickness, no need to have same bevels and no need to have same edge thickness for this particular test (while this thickness bigger then thread thickness).

Hey Vassili. Thanks for your great efforts in the testing you've been doing. I have a question. Could you explain how you know the statement above to be true? Perhaps I missed something in a previous post. Are you saying that if an accurate way to measure sharpness is/can be used, and the measurement is taken every 50 or 100 or whatever cycles just on the very cutting edge, that the overall knife geometry does not play a factor? Thanks for you patience.

Luke

 

[nozh2002]

Hey Vassili. Thanks for your great efforts in the testing you've been doing. I have a question. Could you explain how you know the statement above to be true? Perhaps I missed something in a previous post. Are you saying that if an accurate way to measure sharpness is/can be used, and the measurement is taken every 50 or 100 or whatever cycles just on the very cutting edge, that the overall knife geometry does not play a factor? Thanks for you patience.

Luke

With all do respect... Please, excuse me but I did explain this many time in previous posts as well as many others did, so I'll just continue cutting ropes and threads and kindly ask you to check at leas first page and may be one or two last pages. But to make it clear I suggest you to take manila rope 1/2" thick and cut it about 1/4"-1/2" near the end observing how fibers fall out after being cut and how much of the very edge was really affected - so you may see what role length, width, thickness of the spine and thickness of the edge play in this process.

Thanks, Vassili.

 

[theonew]

Vassili

I'm curious why you choose to use the median value instead of an average or mean value for determining sharpness. Aren't medians generally used to lessen the distortion of excessively low or high numbers when statistically examining a set of numbers? Lets say after 21 thread cuts one knife scores 10 40 gram cuts, 10 50 gram cuts, and 1 60 gram cut, the median being 50. And then lets say another knife scores 1 40 gram cut, 10 50 gram cuts, and 10 60 gram cuts, the median here also being 50. But the average for the first knife would be around 46 and the average for the second knife around 54. Statistics are not my strong suit but I would think with the data you are using taking the mean would yield more accurate results.

 

[Any Cal.]

I believe he went with median to reduce the effect that any fluctuations to an extreme would have on the testing. Using your numbers, theonew, in each of 2 tests, (1) 40g and (1) 60g cut is an extreme measurement, a fluke, statistically speaking. Using the median reduces the weight that the fluke(s) carries on the measurement as a whole. I am just guessing, but you have to remember that Vassili is a statistician, so tries to reduce the variables with numbers.