Wire Edge Prevention

[KnifeGrinders]

I ran my tests this morning, here are the results:

S30V Native 5 Sharpened at 15dps on Wicked Edge

Sharpened Stropped Burred Rolled Burred2 Rolled2
Sample #1 84 81 117 122 129 127
Sample #2 130 118 108 139 94 135
Sample #3 113 121 127 121 129 168
Sample #4 87 98 121 139 127 153
Sample #5 101 126 124 131 154 141
Mean 103 108.8 119.4 130.4 126.6 144.8
Std Dev 17.0 16.8 6.6 7.8 19.1 14.4

Notes:
Sharpened = Straight off the WE, progression 600, 800, 1000, 1500 Dia, Med, Fine Ceramic, 14, 7, 5, 3.5, 1, .5 micron diamond paste on balsa strops.
Stropped = 30 alternating-sides freehand passes on CrOx on leather.
Burred = 50 one-side-only heavy pressure passes on CrOx leather.
Rolled = 5 passes approx 1lb pressure on free-rolling 1/4" brass sleeve (cylinder)
Burred2 = 30 medium-pressure passes on 5 micron diamond paste on leather on WE - same angle as sharpening angle.
Rolled2 = Rolled

Notes:

- I think I must not be hitting the apex on some of my steps (though I raise a detectable burr with all of the stones), overall even my sharpening results were disappointing.
- The Burred2 was visible under 200x magnification (so to your point Mr. Grepper, I don't see advantages at this stage of my knowledge).

Steve, would you be able you to roll a burr-free edge of one of the knives you tested on your brass cylinder, and take a before-after sharpness score?
Without that your numbers reference the rolled burr to itself, i.e. they only tell us that the 5 mcrn burr after-roll scores worse than the 0.5 micron burr - but we do not have a number to tell a burr from the burr-free clean apex.

I have to drop sharpened knives to Mad Mex restaurants this morning, and after that will be able to join your experiments at my end.

[KnifeGrinders]

The test knife is made of ELMAX, HRC 62.
The edge was set at 13 dps on a #1000 CBN, and honed at the edge angle with 5-micron diamonds, followed by 0.5-micron diamonds.
The tip half of the blade was then cleaned of the base burr at a higher angle on felt with 1 micron diamonds, and finished at a higher angle with 0.25 micron diamonds.
This way I got the same blade having clean apex on the tip half, and wire edge on the heel half.

Marked sections on the blade edge, and took the BEFORE sharpness scores in the 3 sections of the tip half and 3 sections of the heel half.
The PT50A sharpness tester average score on the tip half and heel half has a negligible difference of 2 BESS, in other words the sharpness tester cannot detect presence of the wire edge - and as we know, this is typical of hard higher-end steels. Only by wider dispersion of the scores in the heel half one can tell that its apex is not as clean as in the tip half.
I could not take a decent image on my toy microscope, but there is a shiny line along the edge in the heel half that may indicate light dispersion by the wire edge.



For impact I used a 3/4" copper pipe over a linear bearing, the impact assembly weight 153 grams.
Positioned the copper roller perpendicular to the blade.

Rolled the copper roller along the edge 1 cycle only, i.e. once forward and backward, and took sharpness scores AFTER the impact.
The results are encouraging -


Next plan is to test more high-end steels in the similar fashion, and also see if the impact assembly weight can be adjusted not to affect the clean apex at all (if possible at all). Maybe half a rolling cycle, only one roll in one direction, will be adequate.

A regular end user who has any model of the BESS sharpness tester will not need the SET stand, just a simple copper roller of a standard load with a handle of some sort, which is easy to make.

Steve, appreciate your idea and the preliminary testing you've done. We are moving in the right direction with the copper roller.
Thanks mate, Vadim.

[SteveG]

I ran my tests this morning, here are the results:

S30V Native 5 Sharpened at 15dps on Wicked Edge

Sharpened Stropped Burred Rolled Burred2 Rolled2
Sample #1 84 81 117 122 129 127
Sample #2 130 118 108 139 94 135
Sample #3 113 121 127 121 129 168
Sample #4 87 98 121 139 127 153
Sample #5 101 126 124 131 154 141
Mean 103 108.8 119.4 130.4 126.6 144.8
Std Dev 17.0 16.8 6.6 7.8 19.1 14.4

Notes:
Sharpened = Straight off the WE, progression 600, 800, 1000, 1500 Dia, Med, Fine Ceramic, 14, 7, 5, 3.5, 1, .5 micron diamond paste on balsa strops.
Stropped = 30 alternating-sides freehand passes on CrOx on leather.
Burred = 50 one-side-only heavy pressure passes on CrOx leather.
Rolled = 5 passes approx 1lb pressure on free-rolling 1/4" brass sleeve (cylinder)
Burred2 = 30 medium-pressure passes on 5 micron diamond paste on leather on WE - same angle as sharpening angle.
Rolled2 = Rolled

Notes:

- I think I must not be hitting the apex on some of my steps (though I raise a detectable burr with all of the stones), overall even my sharpening results were disappointing.
- The Burred2 was visible under 200x magnification (so to your point Mr. Grepper, I don't see advantages at this stage of my knowledge).

Steve, would you be able you to roll a burr-free edge of one of the knives you tested on your brass cylinder, and take a before-after sharpness score?
Without that your numbers reference the rolled burr to itself, i.e. they only tell us that the 5 mcrn burr after-roll scores worse than the 0.5 micron burr - but we do not have a number to tell a burr from the burr-free clean apex.

I have to drop sharpened knives to Mad Mex restaurants this morning, and after that will be able to join your experiments at my end.

That's what my numbers are trying to do, though I don't really know if my edges are burr-free or not (that's the reason I started this exercise).  It's probably not clear because of the way the data looks unformatted.  Let me restate it here, maybe it will be clearer:

Freshly sharpened blade (presumably no burr): 84, 130, 113, 87, 101
Stropped to refine edge: 81, 118, 121, 98, 126 (didn't seem to do much refining)
Burred (by single-side stropping): 117, 108, 127, 121, 124
Rolled: 122, 139, 121, 139, 131
Reburred (first attempt didn't seem like it created a burr): 129, 94, 129, 127, 154
Re-Rolled: 127, 135, 168, 153, 141

If I missed the point of your message, please feel free to point that out - I'm happy to re-test, but can be a little dense sometimes.

[SteveG]

The test knife is made of ELMAX, HRC 62.
The edge was set at 13 dps on a #1000 CBN, and honed at the edge angle with 5-micron diamonds, followed by 0.5-micron diamonds.
The tip half of the blade was then cleaned of the base burr at a higher angle on felt with 1 micron diamonds, and finished at a higher angle with 0.25 micron diamonds.
This way I got the same blade having clean apex on the tip half, and wire edge on the heel half.

Marked sections on the blade edge, and took the BEFORE sharpness scores in the 3 sections of the tip half and 3 sections of the heel half.
The PT50A sharpness tester average score on the tip half and heel half has a negligible difference of 2 BESS, in other words the sharpness tester cannot detect presence of the wire edge - and as we know, this is typical of hard higher-end steels. Only by wider dispersion of the scores in the heel half one can tell that its apex is not as clean as in the tip half.
I could not take a decent image on my toy microscope, but there is a shiny line along the edge in the heel half that may indicate light dispersion by the wire edge.



For impact I used a 3/4" copper pipe over a linear bearing, the impact assembly weight 153 grams.
Positioned the copper roller perpendicular to the blade.

Rolled the copper roller along the edge 1 cycle only, i.e. once forward and backward, and took sharpness scores AFTER the impact.
The results are encouraging -


Next plan is to test more high-end steels in the similar fashion, and also see if the impact assembly weight can be adjusted not to affect the clean apex at all (if possible at all). Maybe half a rolling cycle, only one roll in one direction, will be adequate.

A regular end user who has any model of the BESS sharpness tester will not need the SET stand, just a simple copper roller of a standard load with a handle of some sort, which is easy to make.

Steve, appreciate your idea and the preliminary testing you've done. We are moving in the right direction with the copper roller.
Thanks mate, Vadim.

Now that's the way a test should be done!  Thanks for doing this, and for the clearly posted results.  That's actually very encouraging.  And not too far from my results; i.e. my average after sharpening was 103, and my average after rolling (and a bunch of other intermediate nonsense) was 144.8.

I guess the good news for me is that my crappy edges probably aren't due to a burr.

Thanks again!

[SteveG]

I ran my tests this morning, here are the results:

S30V Native 5 Sharpened at 15dps on Wicked Edge

Sharpened Stropped Burred Rolled Burred2 Rolled2
Sample #1 84 81 117 122 129 127
Sample #2 130 118 108 139 94 135
Sample #3 113 121 127 121 129 168
Sample #4 87 98 121 139 127 153
Sample #5 101 126 124 131 154 141
Mean 103 108.8 119.4 130.4 126.6 144.8
Std Dev 17.0 16.8 6.6 7.8 19.1 14.4

Notes:
Sharpened = Straight off the WE, progression 600, 800, 1000, 1500 Dia, Med, Fine Ceramic, 14, 7, 5, 3.5, 1, .5 micron diamond paste on balsa strops.
Stropped = 30 alternating-sides freehand passes on CrOx on leather.
Burred = 50 one-side-only heavy pressure passes on CrOx leather.
Rolled = 5 passes approx 1lb pressure on free-rolling 1/4" brass sleeve (cylinder)
Burred2 = 30 medium-pressure passes on 5 micron diamond paste on leather on WE - same angle as sharpening angle.
Rolled2 = Rolled

Notes:

- I think I must not be hitting the apex on some of my steps (though I raise a detectable burr with all of the stones), overall even my sharpening results were disappointing.
- The Burred2 was visible under 200x magnification (so to your point Mr. Grepper, I don't see advantages at this stage of my knowledge).

Steve, would you be able you to roll a burr-free edge of one of the knives you tested on your brass cylinder, and take a before-after sharpness score?
Without that your numbers reference the rolled burr to itself, i.e. they only tell us that the 5 mcrn burr after-roll scores worse than the 0.5 micron burr - but we do not have a number to tell a burr from the burr-free clean apex.

I have to drop sharpened knives to Mad Mex restaurants this morning, and after that will be able to join your experiments at my end.

That's what my numbers are trying to do, though I don't really know if my edges are burr-free or not (that's the reason I started this exercise).  It's probably not clear because of the way the data looks unformatted.  Let me restate it here, maybe it will be clearer:

Freshly sharpened blade (presumably no burr): 84, 130, 113, 87, 101
Stropped to refine edge: 81, 118, 121, 98, 126 (didn't seem to do much refining)
Burred (by single-side stropping): 117, 108, 127, 121, 124
Rolled: 122, 139, 121, 139, 131
Reburred (first attempt didn't seem like it created a burr): 129, 94, 129, 127, 154
Re-Rolled: 127, 135, 168, 153, 141

If I missed the point of your message, please feel free to point that out - I'm happy to re-test, but can be a little dense sometimes.

I just figured out what you meant (told you I can be a bit dense).  Yes, that makes perfect sense, I'll run that test tomorrow (probably superfluous now that you've run yours, but worth the effort).

[KnifeGrinders]

Please do when time permits, and I will also follow my plan.
We interpret the BESS scores the same way you do, and you have a keen understanding of what's happening at the edge through the numbers on the tester display.

As to yours "I don't really know if my edges are burr-free or not (that's the reason I started this exercise)" - we are in the same position, and gradually improve our de-burring technique as we better our burr detection. We are dealing with micro-burrs that cannot be seen or palpated, and my USB microscope can't see them either. Till recently, real cutting was the only way to reveal those burrs - but as this thread goes on we get methods to reveal them, first in mainstream and now in higher-end knives.

[EOU]

Well you've heard of location, location, location in the real estate biz and we think that it is burr, burr, burr in the knife sharpening biz. Thank you to KG and Steve for taking such an in-depth look at the subject. We know that Grepper has spent a considerable amount of time studying this problem as well. No other sharpening flaw creates greater deviation in BESS scores than residual burr. The silver lining here is that this great deviation makes it easy and obvious for your edge tester to detect its presence. Now that we've detected it though, the question of proper removal remains.

We've learned so much about burr in the past year and it is a formidable foe. Removing all the burr while preserving the edge does not seem to be a simple, one size fits all, process to date. Work, like yours, may get us there one day so "hats off" for your efforts.

[SteveG]

Today's results were laughable. I have to go back to work, so don't have time to write it up at the moment, but the bottom line is that my "deburred" edges preformed worse than the ones I tried to force a burr on, so I need to change my methodology. I should have some shop time tomorrow, so I'll repeat the experiment with some modification.

[KnifeGrinders]

Thanks for the report KG. We have a question; your post says "While his own knives last through 4-5 steer carcasses (with steeling), our test knives lasted for 2 carcasses and performed well, though for a shorter period." Did he steel the knives provided by you as well?

Yes.
The knives I gave him are the same as his, only new, and he steeled them as he usually does, using his polished and fine-cut steels after approx. every 10 cuts.
His steeling reflexes are set to a different edge angle though, and with our low angle of 12 dps he must have been oversteeling the apex on our knives, and this might be contributing to them wearing quicker.

Anyway, I know I have to shorten the time the edge contacts the felt wheel rotating at high RPM, and seem to have found a solution by a small modification to the horizontal platform.

We are lucky we met this butcher as he can clearly explain what is wrong and what is good, and we've improved our sharpening working with him.
The very first knives we sharpened for him he discarded in the beginning of a boning session because of the wire edge and started me thinking about it 

So for the next boning our butcher will get 2 identical knives from us: one deburred on a felt wheel run on the half-speed grinder at ~1400 RPM, and the second deburred on the same felt wheel run on Tormek at 90 RPM, not knowing which is what - if the knives show no difference between themselves but dull quicker than his own knives, then it must be due to his steeling.
...

And so today I've sharpened two more identical SWIBO knives for live boning trial - these knives are of the same catalog ID our butcher uses for life.
Both knives are sharpened at 12 dps the same way, with only one difference:

The Knife-1 is deburred on a rock-hard slotted felt wheel run on the half-speed grinder at 1425 RPM


While the Knife-2 is deburred on the same slotted felt wheel run on Tormek at 90 RPM


For this test, I deburred the knives using a knife jig for angle accuracy, to rule out any discrepancies; the honing angle is controlled with our computer software.
Both knives score the same 55 BESS - by what have been discussed in this thread, for mainstream knives this is a strong indicator of a clean apex free of any wire edge.

I've asked our master butcher to compare the knife performance between #1 and #2, and to the similar SWIBO knives he sharpens himself on benchstones - he does not know which knife is what, and can tell them apart only by number of dots on the handle. I will have feedback from the butcher either this Thursday or on Monday.
The outcome will tell us whether deburring on the felt at high RPM degrades the edge retention.

SEE CONTINUATION IN THE THREAD  Overheated edge BESS scores >>
Knives free of wire edge performed every whit as good as knives sharpened by himself

[eddosedge]

Thanks for the report KG. We have a question; your post says "While his own knives last through 4-5 steer carcasses (with steeling), our test knives lasted for 2 carcasses and performed well, though for a shorter period." Did he steel the knives provided by you as well?

Yes.
The knives I gave him are the same as his, only new, and he steeled them as he usually does, using his polished and fine-cut steels after approx. every 10 cuts.
His steeling reflexes are set to a different edge angle though, and with our low angle of 12 dps he must have been oversteeling the apex on our knives, and this might be contributing to them wearing quicker.

Anyway, I know I have to shorten the time the edge contacts the felt wheel rotating at high RPM, and seem to have found a solution by a small modification to the horizontal platform.

We are lucky we met this butcher as he can clearly explain what is wrong and what is good, and we've improved our sharpening working with him.
The very first knives we sharpened for him he discarded in the beginning of a boning session because of the wire edge and started me thinking about it 

So for the next boning our butcher will get 2 identical knives from us: one deburred on a felt wheel run on the half-speed grinder at ~1400 RPM, and the second deburred on the same felt wheel run on Tormek at 90 RPM, not knowing which is what - if the knives show no difference between themselves but dull quicker than his own knives, then it must be due to his steeling.
...

And so today I've sharpened two more identical SWIBO knives for live boning trial - these knives are of the same catalog ID our butcher uses for life.
Both knives are sharpened at 12 dps the same way, with only one difference:

The Knife-1 is deburred on a rock-hard slotted felt wheel run on the half-speed grinder at 1425 RPM


While the Knife-2 is deburred on the same slotted felt wheel run on Tormek at 90 RPM


For this test, I deburred the knives using a knife jig for angle accuracy, to rule out any discrepancies; the honing angle is controlled with our computer software.
Both knives score the same 55 BESS - by what have been discussed in this thread, for mainstream knives this is a strong indicator of a clean apex free of any wire edge.

I've asked our master butcher to compare the knife performance between #1 and #2, and to the similar SWIBO knives he sharpens himself on benchstones - he does not know which knife is what, and can tell them apart only by number of dots on the handle. I will have feedback from the butcher either this Thursday or on Monday.
The outcome will tell us whether deburring on the felt at high RPM degrades the edge retention.

KG, would he be amenable to letting you analyze one of his knives after he sharpens it?

Ed K.