Edge retention as influenced by apex finishing grit

Discuss Spyderco's products and history.
Cliff Stamp
Member
Posts: 3852
Joined: Sat Dec 31, 2005 2:23 pm
Location: Earth
Contact:

Edge retention as influenced by apex finishing grit

#1

Post by Cliff Stamp »

Recently just as part of looking at various abrasives I started doing edge retention trials slicing 1/2" hemp. The knife used is a 3Cr13 stainless steel chef's knife which is sharpened with an edge angle of ~ 5 dps and then micro-beveled at 15 dps with various grit finishes. Each grit finish is checked under magnification. Here is the result of the CBN rods :

Image

As the CBN rods are fairly coarse (400 grit) the apex line is a bit ragged as the scratches are deep. However this is exactly what gives it the high edge retention in slicing abrasive materials and it causes it to blunt in a complicated manner :

-the tips of the jagged bits smooth/round
-the inside of the teeth round out
-finally large flats appear on the edge

Image

This is the edge at the end of the trial, see how the jaggedness is gone and it is not smooth and lightly irregular which is left over from the initial rough finish. At this stage the knife is really dull (sharpness is less than 1.5% of optimal) however it barely reflects light (just barely) and still easily slices newsprint and has long past the point of that high cutting ability you see when the knife is fully sharp and it can slice that 1/2" hemp with 4-5 lbs on a 2" draw.

The results of a bunch of stones :

Image

The curious thing about this is that you could actually use this to estimate the grit of the stones fairly well. This data is only the average of two runs so some of the error bars are quite large. I may repeat them all a few more times to make them more consistent but not necessarily as I did this more to show the general pattern vs exacting specifics so I would rather have a lot of grits than a few which are more precisely determined.

I will eventually add the Sharpmaker diamond, medium and fine rods as I have them as well. Background thread : http://www.cliffstamp.com/knives/forum/read.php?3,37143" target="_blank .
KevinOubre
Member
Posts: 111
Joined: Tue Sep 30, 2014 6:55 pm
Location: Baton Rouge, Louisiana

Re: Edge retention as influenced by apex finishing grit

#2

Post by KevinOubre »

Not to sound like an idiot, but I have some questions about the graph that are not quite clear to me. What does TCE (1.5%)[in] stand for? What is it showing? Also, I am assuming from the initial sharpness measurement of the soft Arkansas, that it is more than 100% of optimal. How is that? Did it exceed a previously established scale you use?
Cliff Stamp
Member
Posts: 3852
Joined: Sat Dec 31, 2005 2:23 pm
Location: Earth
Contact:

Re: Edge retention as influenced by apex finishing grit

#3

Post by Cliff Stamp »

KevinOubre wrote:.... does TCE (1.5%)[in] stand for? What is it showing?
It is an edge retention statistic, it is calculated as the sum of :

(percentage of sharpness) * (length of material cut)

until the knife has reached 1.5% of optimal sharpness. Now you might ask - "Ah, why not just use the number of cuts made in total?" . The answer is hard to explain with out some math or just experience in doing it. In short, that is instable, the way the TCE is calculated it uses all the sharpness data during the cutting trial to produce essentially a sort of "average sharpness" during the cutting.
Did it exceed a previously established scale you use?
The scale I use basically sets 100% at the limit I can achieve with the ideal steel and stone. I keep setting it slightly beyond what I can achieve in a sensible amount of time. This knife is on the far end of being very easy to sharpen and a soft arkansas stone is almost trivial to use to produce a high sharpness. There is nothing really stopping you from getting ultra-high sharpness on any stone or steel, it is just a matter time in most cases.

There is a lot more detail in the thread on the forum about it as I discuss the results for each stone. But just as a point of comparison this is the method I used to set the apex with the CBN rods :

-5 passes with the weight of the knife
-10 passes with full contact, as light as possible
-5 passes with ultra-light force, barely making contact

This however is what I have to do with the King 1000, 50-75 passes using a combination of :

-ultra-light
-cross cutting the scratches
-ultra-high angles
-micro/short stroke
-constantly alternate sides

It takes 5-10 times as long to sharpen in comparions to the CBN rods. There is an interesting question here as well which is what would chart look like if I spent just 15 seconds to set the apex. A lot of these stones would end up producing very low sharpness, most of the waterstones for example aside from the Sigma Power 120.
KevinOubre
Member
Posts: 111
Joined: Tue Sep 30, 2014 6:55 pm
Location: Baton Rouge, Louisiana

Re: Edge retention as influenced by apex finishing grit

#4

Post by KevinOubre »

As always Cliff, thanks for the information. The time question makes a lot of sense, at least for me, as one of the ways I maintain edges between full sharpenings, is using a micro bevel on either my DC4 or Double Stuff pocket stone. Some steels can definitely be problematic but others take to this extremely easily, specific abrasive being used being important as well. I find my Double stuff or DC4 works equally as fast on my high carbide steels like M4 but my DC4 works much better(faster) on steels like 1095, 1075, 5160, etc.
Cliff Stamp
Member
Posts: 3852
Joined: Sat Dec 31, 2005 2:23 pm
Location: Earth
Contact:

Re: Edge retention as influenced by apex finishing grit

#5

Post by Cliff Stamp »

KevinOubre wrote:I find my Double stuff or DC4 works equally as fast on my high carbide steels like M4 but my DC4 works much better(faster) on steels like 1095, 1075, 5160, etc.
The more I play around with abrasives the more I realize that it isn't nearly as trivial as I thought. You can take a stone and a steel and with a certain combination of force and water it works well, but change any of that and it can be really poorly performing.

Here is an interesting way to look at the edge retention :

Image
buckthorn
Member
Posts: 279
Joined: Fri Aug 15, 2008 7:30 pm

Re: Edge retention as influenced by apex finishing grit

#6

Post by buckthorn »

Cliff-thanks for all this fascinating material. In your pie diagram, does each slice represent the retention achieved by using only that abrasive or, as I suspect, is it a diagram of the cumulative retention added by each abrasive when used sequentially? If the latter is the case, why do the slices seem to get progressively smaller except between the 36 dressing and the 120? What's different about the relationship between those first two steps and all the subsequent steps. I hope my question does not display ignorance of what has already been explained in this thread. If, so, I apologize for wasting your time.
Cliff Stamp
Member
Posts: 3852
Joined: Sat Dec 31, 2005 2:23 pm
Location: Earth
Contact:

Re: Edge retention as influenced by apex finishing grit

#7

Post by Cliff Stamp »

buckthorn wrote:does each slice represent the retention achieved by using only that abrasive
Yes.
What's different about the relationship between those first two steps and all the subsequent steps.
Good question.

That 36 grit stone isn't an actual sharpening stone, I used it as a point of curiosity, it is a dressing stone intended to reshape other abrasives, usually power grinding wheels. It has an extremely strong bond and the abrasive isn't very friable and thus it just wears smooth.

Image

The abrasive is simply huge, I can't even shoot it at 50X magnification as all you see is one grit, I have to turn the magnification way down to show the structure of the stone as the grit is so large you can see the individual ones easily by eye :

Image

Now there are two possibilities as to why the edge retention decreases when you go from the 120 Sigma Power to the 36 grit dressing stone :

a) the too large grit itself has become a negative

I can't see immediately why this would be the case, but I just list it as it could be possible.

b) the edge isn't well formed even though it is sharp

This is where it gets a bit technical and it has to do with how very coarse edges blunt. Here is a typical very coarse edge :

Image

The way this blunts is by a sort of staging :

-the tips wear and round
-the inside of the scallops wear
-finally, large flat spots are produced as the tips and scallops are worn away

Here is the worn edge :

Image

While there is still some jaggedness remaining you can see most of it is worn away and the edge has lost most of its sharpness at this point, < 1.5% of optimal.

Now here is what I think is happening with the dressing stone :

Imagine a serrated blade which is fully sharp, now imagine two others :

-one which has fully sharp tips but blunt between the tips
-one which has fully sharp between the tips but the tips are themselves are blunted

If you imagine for example drawing either of those partly sharp blades across a piece of paper then they both will cut it readily because either the tips or the scallops cut it and thus if you just did a quick paper cut with all blades they might appear to be all sharp. The only way you could tell which one was really sharp was to cut something with a small movement and measure the tip sharpness and the scallop sharpness directly.

Now what I think is happening is very similar with the edge left by the dressing stone that it is sharp a little in the same way but not fully sharp as is the edge on the Sigma Power 120. I think this happens because the stone simply doesn't cut well and you can feel a lot of deformation vs abrasion. There is no way to check this directly as it isn't like that serrated blade where you can actually cut something with the tips or scallops as in this case they are all tiny, < 10 microns.

There are also a few other things such as I think the edge on the dressing stone is a bit over strained, again due to deformation. And as always I would want to run all of this data a few times before I would be really confident in the results.

--

At this stage though this is all just a hypothesis, in order to check it out I need to get some other very coarse abrasives and see what happens, where they fall. I have some loose 36 grit Silicon Carbide and I intend to check that and a few others. What would be interesting to see happen would be if the performance took a strong negative past some grit point and the edge retention decreased. I can't immediately explain that so it would be cool to see it happen and they try to figure out why.
Cliff Stamp
Member
Posts: 3852
Joined: Sat Dec 31, 2005 2:23 pm
Location: Earth
Contact:

Re: Edge retention as influenced by apex finishing grit

#8

Post by Cliff Stamp »

Spyderco medium :

Image

My medium rods are quite old (decade+) and are much finer now than they were originally when I initially estimated them at ~12.5 micron : http://www.cliffstamp.com/knives/review ... grits.html" target="_blank . They are among the easiest to use to refine an apex of the stones I have seen :

-do not require soaking
-can not be readily gouged
-very slow wear
-do not rust/corrode
-no slurry formation
-very flat

The fine finish though does produce low edge retention in general doing slicing on abrasive media like this run on 1/2" hemp :

Image

and just the edge retention alone :

Image

Note the huge increase in edge retention slicing hemp when using the CBN rods vs the medium. Even if you were not interested in regrinding knives, this performance factor alone may be of interest depending on what is cut and how.
Henry - get both
Member
Posts: 200
Joined: Wed Jan 04, 2012 10:32 pm

Re: Edge retention as influenced by apex finishing grit

#9

Post by Henry - get both »

Great work Cliff. I would like to see this test with aebl and s30v.
Cliff Stamp
Member
Posts: 3852
Joined: Sat Dec 31, 2005 2:23 pm
Location: Earth
Contact:

Re: Edge retention as influenced by apex finishing grit

#10

Post by Cliff Stamp »

Henry - get both wrote:Great work Cliff. I would like to see this test with aebl and s30v.
I think that is an excellent point because it isn't obvious that these same stones would produce the same results on very different steels. The one I am using now is very easy to grind as it is both fairly soft and very low carbide. If the steel was much harder and had a much higher carbide volume then some of those stones like the Soft Arkansas may not work very well at all as the abrasive (Novaculite) is the same hardness as full hard martensite and much softer than cementite and all alloy carbides.

The other thing would be really interesting as you could ask a similar question as I posed in regards to the edge retention as influenced by apex angle. Lets assume we have this same chart with :

-3Cr13, < 55 HRC
-AEB-L, > 60 HRC
-S30V/Elmax, > 62 HRC
-ZDP-189, M4, > 64 HRC

You could then ask a very interesting question similar to :

-At the same apex angle, what grit is required to sharpen AEB-L to have the same edge retention slicing hemp as S30V sharpened on the medium rods?

Or you could even be much more specific and ask (and answer) a question such as :

-At the same apex angle, what grit is required to sharpen AEB-L to have 50% greater edge retention slicing hemp than S30V sharpened on the medium rods?

The only real problem is one of practicality. This steel, 3Cr13 is both soft and low carbide and has a very basic hardening (the knife cost $1), yet with the very coarse edges it still cuts ~1000 pieces of 1/2" hemp in a run and it still slices newsprint at the end of that amount. With steels which do this kind of work inherently better, then this amount of rope will climb and it could take far more pieces of hemp to blunt them at the very low grit levels.

I think I will do an AEB-L run, but I am not sure about an Elmax/S30V run simply because it would literally require on the order of a quarter of a million cuts into 1/2" hemp. There is a time/cost consideration there. But I do agree it is a very interesting question.
Henry - get both
Member
Posts: 200
Joined: Wed Jan 04, 2012 10:32 pm

Re: Edge retention as influenced by apex finishing grit

#11

Post by Henry - get both »

Cliff Stamp wrote: I think that is an excellent point because it isn't obvious that these same stones would produce the same results on very different steels.
So true. 3Cr13 seems to require such a coarse edge to have decent edge retention on manilla rope. I don't think aebl at 60rc would be so fickle.

Its interesting, looking at the grit / edge retention chart, if I were to guess the grit of the DMT MXF based on its edge retention I would guess it was 1500 grit. Seems like an outlier.
arty
Member
Posts: 416
Joined: Sat Oct 09, 2010 11:59 am

Re: Edge retention as influenced by apex finishing grit

#12

Post by arty »

Isn't this a low carbon steel knife? I wouldn't expect it to do any better than the sort of steel used in a butter knife, or am I missing something?
Cliff Stamp
Member
Posts: 3852
Joined: Sat Dec 31, 2005 2:23 pm
Location: Earth
Contact:

Re: Edge retention as influenced by apex finishing grit

#13

Post by Cliff Stamp »

Henry - get both wrote:
So true. 3Cr13 seems to require such a coarse edge to have decent edge retention on manilla rope. I don't think aebl at 60rc would be so fickle.
I think in fact this is a critical issue. If you look at the above the performance of that knife on the hemp at the far left where it is the highest is many times to one greater than the K2/Farid in 10V at 63 HRC. How do you thus describe the edge retention of the steel? Would you be willing to sharpen such a knife with the same edge/grit because if you don't then this knife has better edge retention on slicing hemp and similar materials.

However I have been considering a different metric, what about if instead of looking at edge retention just on slicing, you looked at this number :

-TCE (slicing) * TCE (push cutting)

If you take the 120 Sigma Power then the push cutting performance would be very low so that metric which is a kind of average edge retention over normal slicing and push cutting would be low as well. However if you move up to AEB-L then you can get higher edge retention at a finer finish and this might allow a greater total maximum performance. At this point I don't have the data, but it seems reasonable. In this way you could say then that AEB-L has a higher edge retention in an absolute sense over a range of grits and apex angles.



Its interesting, looking at the grit / edge retention chart, if I were to guess the grit of the DMT MXF based on its edge retention I would guess it was 1500 grit. Seems like an outlier.
It looks to be so, but that is the problem with forcing outliers. If you think you know the pattern and just chop off the data that don't fit then you obviously produce the pattern, but how do you know it isn't a real measurement. I suspect that it is for a few reasons. The main one is that I think that the general trend will not hold as you go to finer finishes, in fact I think it will stop decreasing and then will start to increase again. This will happen I believe because the blunting will change in manner as the cuts become more of a push cut even if you attempt a draw. I am going to run the Spyderco fine and few other very high polish finishes to check this out.
Cliff Stamp
Member
Posts: 3852
Joined: Sat Dec 31, 2005 2:23 pm
Location: Earth
Contact:

Re: Edge retention as influenced by apex finishing grit

#14

Post by Cliff Stamp »

arty wrote:Isn't this a low carbon steel knife? I wouldn't expect it to do any better than the sort of steel used in a butter knife, or am I missing something?
It is a very low carbon, low carbide and fairly soft knife, < 55 HRC, it can be easily filed. It is however still a very different steen from what is found in pots and pans and general flatware which are not martensitic steels. It is similar to 420, AUS-4, 1050 and similar steels. In regards to the performance, for reference, the K2/Farid in 10V at 63 HRC (based on Phil's measurements on his blade) with the as-boxed edge/sharpness, only has a TCE of ~3 in, thus this blade has higher edge retention on the CBN finishes and the others to the left of it. The point of this comparison is to show the influence of various stones on edge retention, the same thing is seen on all steels (though it won't be identical). I chose this knife for a number of reasons, one of them being it is going to be heavily consumed in it as this work is going to require it to be sharpened ~150 times and some of those times are with very coarse stones. Plus there would be other practical considerations of time/cost if I used AEB-L or Elmax and it would not gain more information at this stage though I might do isolated runs with those steels later with some of the abrasives.
Henry - get both
Member
Posts: 200
Joined: Wed Jan 04, 2012 10:32 pm

Re: Edge retention as influenced by apex finishing grit

#15

Post by Henry - get both »

Yea, push cutting manilla rope with 3cr13 should be a brief test. I would have a good laugh if I saw someone push cutting manilla rope while not doing a specific push cut test. Maybe I would ask them why they were trying to mash their edge with rope instead of with a rock.
Cliff Stamp
Member
Posts: 3852
Joined: Sat Dec 31, 2005 2:23 pm
Location: Earth
Contact:

Re: Edge retention as influenced by apex finishing grit

#16

Post by Cliff Stamp »

Henry - get both wrote:I would have a good laugh if I saw someone push cutting manilla rope while not doing a specific push cut test.
This knife cuts that rope on a straight push very easily with a high polish, it only requires ~5 lbs even with a 1/2 piece of hemp, on smaller ropes it just falls right through. It is very thin stock, edge is ~5 dps and the micro-bevel is < 0.001" thick. With a very high polish it has little benefit to make a draw cut as there is almost no cutting horizontally as there is essentially nothing to make any kind of sawing effect as the edge is very smooth. This is a polished edge from a ~5 micron waterstone at 50X magnification :

Image

Care has to be taken when talking about polished edges on rope and similar as people often have the idea that a polished edge slips on rope, well that knife simply isn't sharpened well and is likely over buffed and rounded. If the knife is sharpened well then as soon as force is put into the knife it should cut into the rope immediately and if you try to draw the knife it should just fall through the rope. However the over promotion of stropping, especially with lots of force, soft backing and just many, many passes has many people think that polished edges can't actually cut rope.

Image

I added a few high polishes and you can see that the general trend of decreasing performance stops at some point and starts to rebound. The reason for this is because the cutting mechanic, and thus the blunting mechanic changes completely. It should be obvious that the decrease in performance could not continue as the immediate from from the 120 Sigma Power would predict because if you use knives with a very high polish it is obvious they don't go blunt immediately. An interesting question to me is how high will the performance rebound up and what does this number look like :

TCE (slicing) * TCE (push)

where would the maximum of that peak be as it would be an interesting case for the best general purpose type edge. I tend to use 600 DMT to MXF DMT frequently and would guess just roughly at this stage from what I have seen in just general use, that the maximum would be between them and closer to the MXF side.
Henry - get both
Member
Posts: 200
Joined: Wed Jan 04, 2012 10:32 pm

Re: Edge retention as influenced by apex finishing grit

#17

Post by Henry - get both »

Well I look forward to seeing the push cut edge retention results with 3Cr13 finished on the sigma 120.
User avatar
Brock O Lee
Member
Posts: 3297
Joined: Thu Jul 21, 2011 3:34 am
Location: Victoria, Australia

Re: Edge retention as influenced by apex finishing grit

#18

Post by Brock O Lee »

This is very useful info, especially in graph form, thanks for the effort.

I am used to Spyderco Medium micro bevels, which ranks the lowest on your list. This gives me more than enough reason to get those fine and extra fine DMT's I've been looking at. Small investment for a potential large pay-off.

Would you say the Spyderco Diamond rods and CBN rods are comparable in grit?

What grit would you say is a good choice to apex with to form a clean crisp edge, before you apply the micro?
Hans

Favourite Spydies: Military, PM2, Shaman, UKPK
Others: Victorinox Pioneer, CRK L Sebenza 31, CRK L Inkosi
Mallus
Member
Posts: 151
Joined: Wed Oct 24, 2007 8:02 am

Re: Edge retention as influenced by apex finishing grit

#19

Post by Mallus »

Hi Cliff,

Very interesting work, as always!

I wonder if at some point you'd be able to put together a webpage that would list all(?) the tested parameters, offer condensed summaries of the results and link to where the complete work is presented?

I understand this is rather a time consuming job, but I think it would be useful in two ways. First, it would benefit those who have a a hypothesis about something knife related to check whether it's done already and what was your take on it. Second, it would be nice to be able to guide people to a location of solid scientific knowledge, when the same funny myths pop up in the knife discussions again and again. In the end, this kind of arrangement might even save your time when instead of explaining the same things repeatedly, you could in many cases just link to one page, which would likely help answering a few follow up questions also.
Laethageal
Member
Posts: 543
Joined: Wed Jan 23, 2013 6:24 pm
Location: Lost in my thoughts

Re: Edge retention as influenced by apex finishing grit

#20

Post by Laethageal »

He's got his own forum where you can find most of this and then some more. Ain't a webpage that got it all in one place, but if you wanna check it out:
http://www.cliffstamp.com/knives/forum/" target="_blank
If it's not polished, call it a saw, not an edge!
Post Reply