Thread: experiments

Thank you for your reply Bart.

What I was referring to by animosity was the sudden turn to accusation of bias, fraud, ignorance. This in spite of the fact the description of the method specifically mentions the aspects of lighting and limitations of the equipment. Ignorant statements about how the appearance of scratches can be manipulated to appear smooth or deeply furrowed, combined with assurance that these cannot be 30k because they are too coarse in appearance is far less than I expect from an intelligent conversation.

This kind of response, and I have seen it before. Along with the mindless buddy system of thanking up the rep, like some middle school popularity contest truly makes me question my participation here.

I just want to check something ... you guys do realise that his image size is one centimetre NOT one millimetre don't you?

He mislabelled it in at least one place:

Originally Posted by Rajagra

I just want to check something ... you guys do realise that his image size is one centimetre NOT one millimetre don't you?

He mislabelled it in at least one place:

yes. As I've stated several times in this thread, those photos don't demonstrate what that poster claims they demonstrate.

Originally Posted by kevint

Thank you for your reply Bart.

What I was referring to by animosity was the sudden turn to accusation of bias, fraud, ignorance. This in spite of the fact the description of the method specifically mentions the aspects of lighting and limitations of the equipment. Ignorant statements about how the appearance of scratches can be manipulated to appear smooth or deeply furrowed, combined with assurance that these cannot be 30k because they are too coarse in appearance is far less than I expect from an intelligent conversation.

This kind of response, and I have seen it before. Along with the mindless buddy system of thanking up the rep, like some middle school popularity contest truly makes me question my participation here.

His photos do not demonstrate what he claims they demonstrate. To see submicron features requires a microscope capable of a very high magnification. There have been multiple papers published (Verhoeven's Knife Sharpening Experiments 2002 paper, and the Popular Mechanics article from the 20's) that show these scratches accurately, and both of these used 3000x magnification. Verhoeven used an electron microscope, and the PM article used a custom-built optical microscope.

The 25x magnification the linked poster used is simply inadequate by roughly two orders of magnitude. It is not ignorance nor mindlessness to point this out, nor is it ignorance nor a matter of popularity to note that this means that the linked poster is either unaware that he needs a much bigger scope or he is intentionally trying to deceive the readers. And it seems reasonable to point out that one factor that makes the ignorance option more likely (or at least more understandable) is the possibility that confirmation bias may have played a part. And it is not unintelligent to note that the sorts of effects that the photos do show are the sort that have tripped up previous attempts to analyse hone scratches with low-power microscopes. Nor is it out of line to describe how small differences in lighting can have a huge effect on the sorts of surface features that show up in the photos.

It is possible that the OP of the linked thread is absolutely correct in his conclusion that the Nakayama hone is markedly superior to the Shapton 30k. I've got both and think they're roughly equivalent, but it's possible that I'm wrong. But this conclusion is not bolstered by the photos in that linked article. I'm sorry to disappoint you, but the resolution is just too low. Even at 3000x the scratches are surprisingly small, and at 3000x you can look at the blade edge-on and measure the width of the cutting edge to the hundredth of a micron, and at that magnification the scratches are roughly the same size as the ones shown in the 30k shapton photo. So if the guy in the linked thread is right about those scratches, then somebody needs to tell Prof John Verhoeven at the University of Iowa that he needs to learn how to use an electron microscope. Or maybe, just maybe, it's the random internet poster with the 25x microscope that's wrong.

Maybe.

Originally Posted by Rajagra

I just want to check something ... you guys do realise that his image size is one centimetre NOT one millimetre don't you?

He mislabelled it in at least one place:

right. as labeled: 1/2 mm scale. I may have read over a typo, but I thought it said. 1cm was enlarged to to fit 25cm screen which equaled an approximate 25X magnification.

Originally Posted by kevint

right. as labeled: 1/2 mm scale. I may have read over a typo, but I thought it said. 1cm was enlarged to to fit 25cm screen which equaled an approximate 25X magnification.

Yes, each division on that ruler is 0.5mm, total widtch about 1cm. Where he says "Width of Sample Equals 1mm" it is wrong.

I only mention it because I've been trying to take some pictures through a microscope and it's difficult. My jaw dropped when I saw the quality of his pictures until I realised they weren't really micrographs at all.

Here are my first feeble attempts, scale roughly 1mm top to bottom,
Unbranded Double Arrow as bought, Dovo honed by me (after shaving, unstropped), Wapi bought shave-ready (used & stropped.):

Originally Posted by mparker762

His photos do not demonstrate what he claims they demonstrate. To see submicron features requires a microscope capable of a very high magnification. There have been multiple papers published (Verhoeven's Knife Sharpening Experiments 2002 paper, and the Popular Mechanics article from the 20's) that show these scratches accurately, and both of these used 3000x magnification. Verhoeven used an electron microscope, and the PM article used a custom-built optical microscope.

The 25x magnification the linked poster used is simply inadequate by roughly two orders of magnitude. It is not ignorance nor mindlessness to point this out, nor is it ignorance nor a matter of popularity to note that this means that the linked poster is either unaware that he needs a much bigger scope or he is intentionally trying to deceive the readers. And it seems reasonable to point out that one factor that makes the ignorance option more likely (or at least more understandable) is the possibility that confirmation bias may have played a part. And it is not unintelligent to note that the sorts of effects that the photos do show are the sort that have tripped up previous attempts to analyse hone scratches with low-power microscopes. Nor is it out of line to describe how small differences in lighting can have a huge effect on the sorts of surface features that show up in the photos.

It is possible that the OP of the linked thread is absolutely correct in his conclusion that the Nakayama hone is markedly superior to the Shapton 30k. I've got both and think they're roughly equivalent, but it's possible that I'm wrong. But this conclusion is not bolstered by the photos in that linked article. I'm sorry to disappoint you, but the resolution is just too low. Even at 3000x the scratches are surprisingly small, and at 3000x you can look at the blade edge-on and measure the width of the cutting edge to the hundredth of a micron, and at that magnification the scratches are roughly the same size as the ones shown in the 30k shapton photo. So if the guy in the linked thread is right about those scratches, then somebody needs to tell Prof John Verhoeven at the University of Iowa that he needs to learn how to use an electron microscope. Or maybe, just maybe, it's the random internet poster with the 25x microscope that's wrong.

Maybe.

Mr. Parker. I considered the tone of replies to be quite negative. Your text overall; contradictory.

Originally Posted by kevint

Mr. Parker. I considered the tone of replies to be quite negative. Your text overall; contradictory.

Please, I fouind Mr. Parker's tone very much in line with yours to which I think he was responding. I found the content to be interesting, informative, and in my opinion, quite valid.
"This kind of response, and I have seen it before. Along with the mindless buddy system of thanking up the rep, like some middle school popularity contest truly makes me question my participation here."

So far the discussion has been directly to the topic. The analysis of the study, done by someone else, which you presented. I think it's fascinating, pertinent, and very worth dissecting. There's nothing wrong with attacking the data and/or methods used to reach a conclusion. I do not mean this to be inflammatory, but what did you find contradictory about Mr. Parker's text? If it was content, then please elaborate. If it was tone, then hopefully we can just let it drop.

(Don't let this valuable thread die over this.)

Finally found the Verhoeven paper, which should help correct the overemotional rhetoric and ad-hominem attacks in this thread.


Experiments on Knife Sharpening
September 2004
Prof. John Verhoeven
Emeritus Professor
Department of Materials Science and Engineering
Iowa State University

Experiments on Knife Sharpening


On page 19, figure 23, he shows a blade sharpened on a 6000 grit waterstone, at 3000x and 800x resolutions. Notice how small the grooves are, even at 800x magnification.

On page 30, figure 37, he shows a blade with a secondary bevel formed on chrome oxide. He doesn't give the grit, but he does describe it as the standard chrome oxide block sold in woodworking stores for honing chisels. At best this is the 0.5 micron Lee Valley block, at worst it's the coarse stuff that's blended with other abrasives. If it's the 0.5 micron chrome oxide then the grit is comparable to the grit of a 30k Shapton (my Shapton is 0.48 micron). Even at 3000x, the grooves in the secondary bevel are very thin and very shallow, and are therefore unlikely to be visible at 25x as the author of that linked article claims. Admittedly I'm comparing a 0.5 micron paste to a 0.5 micron stone, but Verhoeven shows an awful lot of photos of 1k, 6k, and 8k honing marks at a variety of resolutions, and the marks from even these relatively coarse hones are surprisingly small.

I do not apologize for being negative about the linked post, nor for being contradictory about his assertions. But if Prof. Verhoeven's pictures are accurate depictions of micron and submicron-scale honing marks, then it is clear that the centimeter-scale photos in the linked thread must be showing much larger honing marks and not the 0.5 micron - 0.2 micron striations that the Shapton and Nakayama hones would be producing.

I hope I haven't soured you on the Verhoeven paper, but it is well worth reading front to back. He shows comparative micrographs of commercial Gillette razor blades and a straight razor that was honed by somebody that used it, and the straight razor comes off looking quite well. And the photos are quite eye-popping.

One other issue with that guy's honing is that he is simply polishing the bevel, not getting the egde any sharper!

Check it out. The bevel is certainly seeing what seems to be improvement, but the dark line along the edge remains untouched.

Good example of what happens if you do not "set your bevel" and move up the progression anyhow. You get nice shiny bevels, and a razor that still won't cut right...